Upload
others
View
14
Download
0
Embed Size (px)
Citation preview
i
Inês Alexandra Manata Antunes Valente
Licenciada em Ciências da Engenharia Química e Bioquímica
Adsorption equilibria of flue gas components on activated
carbon
Dissertação para obtenção do Grau de Mestre em
Engenharia Química e Bioquímica
Orientador: Doutora Isabel Alexandra de Almeida Caneto Esteves Esperança
Co-orientadores: Professor Doutor José Paulo Barbosa Mota
Doutor Rui Pedro Pinto Lopes Ribeiro
Março 2014
Monte da Caparica, Faculdade de Ciências e Tecnologia, Universidade de Lisboa
ii
iii
Adsorption equilibria of flue gas components on activated
carbon
Copyright ©
Eu, Inês Alexandra Manata Antunes Valente, declaro que a Faculdade de
Ciências e Tecnologia e a Universidade Nova de Lisboa têm o direito, perpétuo e sem
limites geográficos, de arquivar e publicar esta dissertação através de exemplares
impressos reproduzidos em papel ou de forma digital, ou por qualquer outro meio
conhecido ou que venha a ser inventado, e de a divulgar através de repositórios
científicos e de admitir a sua cópia e distribuição com objectivos educacionais ou de
investigação, não comerciais, desde que seja dado crédito ao autor e editor.
iv
v
Acknowledgments
No limiar do final de mais uma etapa deste caminho de longos anos, não podia deixar
passar a oportunidade de agradecer às pessoas que me acompanharam até aqui e aquelas
que se destacaram no desfecho deste Mestrado. Gostaria então de agradecer:
Ao Professor Doutor José Paulo Mota e à Doutora Isabel Esteves pela orientação e,
por me terem convencido a fazer a minha tese em Adsorção Gasosa, um tema ao qual me
afeiçoei e do qual tantos conhecimentos adquiri.
Ao Doutor Rui Ribeiro, pela orientação, paciência e disponibilidade contínua para as
minhas “questões mais esotéricas”.
À Faculdade de Ciências e Tecnologia, da Universidade Nova de Lisboa, local que
tanto contribuí-o para o meu desenvolvimento pessoal e profissional.
Ao Professor Mário Eusébio pelos seus conselhos.
Ao Doutor Ricardo Silva pela boa disposição e ajuda na parte informática. À Eliana
Orfão, pela sua constante alegria e positivismo.
Aos meus colegas, vocês que partilharam comigo as dificuldades destes últimos
meses: João Gomes, pela companhia, boa disposição e apoio; e à Bárbara Camacho, pela
orientação, disponibilidade e amizade.
Ao Nuno Costa, à Inês Matos e à Professora Isabel Fonseca que tanto me ajudaram no
capítulo de caracterização do material.
À Dª Palminha e à Dª Maria José Carapinha, pela vossa ajuda.
Aos meus amigos, ao pessoal do BEST Almada, às/aos minhas/meus colegas da
Duplix | Impressão e Imagem pela vossa força, apoio e compreensão.
Aos meus pais, avós, irmão e ao Diogo pelo vosso apoio e amor incondicional para o
qual não existem palavras para descrever.
A todos vós, muito obrigada!
Inês Valente
vi
vii
“Attitude is a little thing that makes a big difference.”
Winston Churchill
Gostaria de dedicar esta tese aos meus pais e avós, que
sempre fizeram de tudo para me darem a melhor educação.
viii
ix
Abstract
Carbon dioxide (CO2) is the greenhouse gas which can be found at higher
concentrations in the atmosphere. This is mainly due to emission of CO2 from anthropogenic
sources as the flue gases fossil fueled power stations.
Adsorption processes are considered as a viable alternative to perform the capture of
the CO2 emitted from flue gases. The development of adsorption-based technologies depends
on the knowledge of the adsorption equilibrium properties of the flue gas components over
potential adsorbent materials This work consisted in the characterization of two activated
carbons: ANGUARD 6, 1 mm, in the form of extruded (Sutcliffe Speakman Carbons Ltd., UK)
and a honeycomb monolith (Mast Carbon International Limited, UK). Surface chemistry
characterization of both carbons was performed. Characterization of the surface area, pore
volumes and pore size distribution was also performed for the ANGUARD 6 sample.
Adsorption equilibrium of carbon dioxide (CO2), nitrogen (N2) and butane (C4H10) at
303.15K, 323.15K and 353.15K in a pressure range of 0-35 bar was measured on ANGUARD
6. Adsorption equilibrium of CO2 on the activated carbon honeycomb monolith was also
measured in the same temperature and pressure ranges as for the ANGUARD 6 sample. The
Sips isotherm model was employed to fit the experimental data and the model could fit the data
successfully. The isosteric heats of adsorption for each of the studied species were also
determined.
Keywords: adsorption equilibrium, activated carbon, carbon dioxide, gravimetric
method.
x
xi
Resumo
O dióxido de carbono é um gás com efeito de estufa que pode ser encontrado em
maiores concentrações na atmosfera. Este facto deve-se, principalmente, a emissões de
origem antropogénica nas quais se inclui a emissão de gases de chaminé de centrais de
produção de energia a partir de combustíveis fósseis.
Processos de adsorção são considerados como uma opção viável para aplicação na
captura de CO2 de gases de chaminé. O desenvolvimento de processos de separação por
adsorção depende no conhecimento das propriedades de equilíbrio de adsorção dos
componentes dos gases de chaminé por potenciais adsorventes.
Este trabalho consistiu na caracterização de dois carvões activados: ANGUARD 6, 1
mm, em forma de extrudados (Sutcliffe Speakman Carbons Ltd., UK) e um monólito de
estrutura tipo “favo de mel” (Mast Carbon International Limited, UK). A química de superfície de
ambos os carvões foi caracterizada. Caracterização da área superficial, volume de poros e
distribuição de tamanho de poros foi efectuada para a amostra de ANGUARD 6.
Foi estudado o equilíbrio de adsorção de dióxido de carbono (CO2), azoto (N2) e butano
(C4H10) a temperaturas de 303.15K, 323.15K e 353.15K, na gama de pressão de 0 a 35 bar, na
amostra de ANGUARD 6. Equilíbrio de adsorção de CO2 no monólito de carvão activado foi
também estudado, na mesma gama de pressão e temperatura. O modelo de isotérmica de Sips
foi utilizado para descrever os dados obtidos experimentalmente. Os calores isostéricos dos
vários adsorbatos estudados foram também determinados.
Palavras-chave: equilíbrio de adsorção, carvão activado, dióxido de carbono, método
gravimétrico.
xii
xiii
Contents 1. Introduction ........................................................................................................................ 1
1.1. Motivation .................................................................................................................. 1
1.2. Thesis Structure ........................................................................................................ 2
Chapter 2 ....................................................................................................................................... 5
2. Background ....................................................................................................................... 5
2.1. Adsorption ................................................................................................................. 5
2.2. Adsorbates .............................................................................................................. 10
2.3. Adsorbents .............................................................................................................. 11
Chapter 3 ..................................................................................................................................... 17
3. Adsorbent Characterization ............................................................................................. 17
3.1. Introduction .............................................................................................................. 17
3.2. Characterization Methods ....................................................................................... 18
3.3. Summary ................................................................................................................. 34
Chapter 4 ..................................................................................................................................... 37
4. Adsorption Equilibrium .................................................................................................... 37
4.1. Introduction .............................................................................................................. 37
4.2. Experimental Description......................................................................................... 37
4.3. Experimental Results and Data Analysis ................................................................ 42
4.4. Summary ................................................................................................................. 64
Chapter 5 ..................................................................................................................................... 67
5. Conclusions and Suggestions for Future Work ............................................................... 67
5.1. Conclusions ............................................................................................................. 67
5.2. Suggestions for Future Work ................................................................................... 69
References .................................................................................................................................. 71
APPENDIX .................................................................................................................................. 75
A. Results from Chapter 3 ....................................................................................................... 76
A.1. Calculus used in the analysis of Bohem Titrations Results ......................................... 76
A.2. Results from Bohem Titrations ..................................................................................... 77
A.3. Results from N2 adsorption at 77K ............................................................................... 79
A.4. Results from BET Surface Area Method Analysis ....................................................... 80
A.5. Results from t-Plot Method Analysis ............................................................................ 80
A.6. Results from Horvath-Kawazoe (HK) Method Analysis ............................................... 81
A.7. Results from Density Functional Theory (DFT) Method Analysis ................................ 81
A.8. Results from Mercury Porosimetry Analysis ................................................................ 81
A.9. Resume of the physical parameters calculated from the several characterization
methods for ANGUARD 6 ................................................................................................... 82
A.10. Calculus used for the determination of the bulk densities for ANGUARD 6 and ACHM
............................................................................................................................................. 82
xiv
B. Results from Chapter 4 ....................................................................................................... 85
C. Equipment Description .................................................................................................... 95
C.1. Equipment Description for Adsorption Equilibrium ...................................................... 95
C.2. Bank of Images ............................................................................................................ 99
xv
List of Figures
Figure 2.1 - IUPAC gas physisorption isotherm classification [13]. .............................................. 7
Figure 2.2 - Generic volumetric apparatus [17]. ............................................................................ 9
Figure 2.3 - Generic gravimetric apparatus [17]. ......................................................................... 10
Figure 2.4 - Hexagonal structure of graphite [13]........................................................................ 14
Figure 2.5 - Schematic representation of an activated carbon porous matrix [42]. .................... 15
Figure 3.1 - Activated carbons used in this study: ANGUARD 6 (on the left) and activated
carbon honeycomb monolith (ACHM). ........................................................................................ 17
Figure 3.2 - Simplified schematic of some acidic surface groups on an activated carbon [28]. . 20
Figure 3.3 - Schematic of some possible basic groups on an activated carbon [28]. ................. 20
Figure 3.4 – TGA analysis of ANGUARD 6. ................................................................................ 25
Figure 3.5 - Adsorption isotherm of N2 at 77K for ANGUARD 6. ................................................ 26
Figure 3.6 - Micropore size distribution obtained from Horvath-Kawazoe Method for ANGUARD
6. .................................................................................................................................................. 31
Figure 3.7 - Pore size distribution obtained from DFT Method for ANGUARD 6. ....................... 32
Figure 3.8 - Experimental mercury intrusion-extrusion cycle for ANGUARD 6. .......................... 33
Figure 3.9 - Illustration of bulk, apparent and skeletal densities [58]. ......................................... 34
Figure 4.1 - Magnetic suspension balance components [60]. ..................................................... 38
Figure 4.2 - Schematic diagram of the experimental apparatus used in the equilibrium
measurements. ............................................................................................................................ 40
Figure 4.3 - Blank calibration of sample holder #1 used in the adsorption, using helium at
293.63K. ...................................................................................................................................... 43
Figure 4.4- Blank calibration of sample holder #2 used in the adsorption, using helium at
293.63K. ...................................................................................................................................... 44
Figure 4.5 - Helium measurements on ANGUARD 6 (top) and ACHM (bottom) at 353.15K...... 45
Figure 4.6 - Net (◊), excess (□) and total (∆) amount adsorbed for nitrogen (N2) at 303.15K for
ANGUARD 6. The solid symbols represent adsorption and the open desorption. ..................... 47
Figure 4.7 - Sips model fitting of the N2 experimental data at 303K, 323K and 353K on
ANGUARD 6 and parameters obtained. Symbols represent the experimental data and the
surface is the global isotherm model. .......................................................................................... 49
Figure 4.8 - Single component N2 isotherms at 303K, 323K and 353K on the activated carbon
ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption; empty
symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for
303.15K, 323.15K, and 353.15K are 6.98, 9.01 and 3.55, respectively. The N2 overall ARE error
is 6.51%. ...................................................................................................................................... 50
Figure 4.9 – Logarithmic representation of the single component N2 isotherms at 303K, 323K
and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data (filled
symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips
model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 6.98, 9.01 and 3.55,
respectively. The N2 overall ARE error is 6.51%. ........................................................................ 50
Figure 4.10 - Sips model fitting of the C4H10 experimental data at 303K, 323K and 353K on
ANGUARD 6 and parameters obtained. Symbols represent the experimental data and the
surface is the global isotherm model. .......................................................................................... 51
Figure 4.11 - Single component C4H10 isotherms at 303K, 323K and 353K on the activated
carbon ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption;
empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE
errors for 303.15K, 323.15K, and 353.15K are 3.90, 6.38 and 3.62, respectively. The C4H10
overall ARE error is 4.63%. ......................................................................................................... 51
Figure 4.12 - Logarithmic representation of the single component C4H10 isotherms at 303K,
323K and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data
(filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the
xvi
Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 3.90, 6.38 and 3.62,
respectively. The C4H10 overall ARE error is 4.63%. .................................................................. 52
Figure 4.13- Sips model fitting of the CO2 experimental data at 303K, 323K and 353K on
ANGUARD 6 and parameters obtained. Symbols represent the experimental data and the
surface is the global isotherm model. .......................................................................................... 52
Figure 4.14 - Single component CO2 isotherms at 303K, 323K and 353K on the activated
carbon ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption;
empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE
errors for 303.15K, 323.15K, and 353.15K are 6.10, 6.20 and 8.45, respectively. The CO2
overall ARE error is 6.92%. ......................................................................................................... 53
Figure 4.15 - Logarithmic representation of the single component CO2 isotherms at 303K, 323K
and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data (filled
symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips
model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 6.10, 6.20 and 8.45,
respectively. The CO2 overall ARE error is 6.92%. ..................................................................... 53
Figure 4.16 - Sips model fitting of the CO2 experimental data at 303K, 323K and 353K on ACHM
and parameters obtained. Symbols represent the experimental data and the surface is the
global isotherm model. ................................................................................................................ 54
Figure 4.17 - Single component CO2 isotherms at 303K, 323K and 353K on the activated
carbon ACHM. Symbols represent the experimental data (filled symbol – adsorption; empty
symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for
303.15K, 323.15K, and 353.15K are 5.29, 4.33 and 6.49, respectively. The CO2 overall ARE
error is 5.37%. ............................................................................................................................. 54
Figure 4.18 - Logarithmic representation of the single component CO2 isotherms at 303K, 323K
and 353K on the activated carbon ACHM. Symbols represent the experimental data (filled
symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips
model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 5.29, 4.33 and 6.49,
respectively. The CO2 overall ARE error is 5.37%. ..................................................................... 55
Figure 4.19 - Single-component isosteric heat of adsorption for N2 at 303.15K, 323.15K and
353.15K as a function of fractional loading on the activated carbon ANGUARD 6, predicted by
Sips isotherm model. ................................................................................................................... 56
Figure 4.20 - Single-component isosteric heat of adsorption for N2 at 303.15K, 323.15K and
353.15K as a function of equilibrium pressure on the activated carbon ANGUARD 6, predicted
by Sips isotherm model. .............................................................................................................. 56
Figure 4.21 - Single-component isosteric heat of adsorption for C4H10 at 303.15K, 323.15K and
353.15K as a function of fractional loading on the activated carbon ANGUARD 6, predicted by
Sips isotherm model. ................................................................................................................... 57
Figure 4.22 - Single-component isosteric heat of adsorption for C4H10 at 303.15K, 323.15K and
353.15K as a function of equilibrium pressure on the activated carbon ANGUARD 6, predicted
by Sips isotherm model. .............................................................................................................. 57
Figure 4.23 - Single-component isosteric heat of adsorption for CO2 at 303.15K, 323.15K and
353.15K as a function of fractional loading on the activated carbon ANGUARD 6, predicted by
Sips isotherm model. ................................................................................................................... 58
Figure 4.24 - Single-component isosteric heat of adsorption for CO2 at 303.15K, 323.15K and
353.15K as a function of equilibrium pressure on the activated carbon ANGUARD 6, predicted
by Sips isotherm model. .............................................................................................................. 58
Figure 4.25 - Single-component isosteric heat of adsorption for CO2 at 303.15K, 323.15K and
353.15K as a function of fractional loading on the activated carbon ACHM, predicted by Sips
isotherm model. ........................................................................................................................... 59
Figure 4.26 - Single-component isosteric heat of adsorption for CO2 at 303.15K, 323.15K and
353.15K as a function of equilibrium pressure on the activated carbon ACHM, predicted by Sips
isotherm model. ........................................................................................................................... 59
xvii
Figure 4.27 - Single-component adsorption isotherms of N2, C4H10 and CO2 on ANGUARD 6 at
303.15 K. The symbols represent the experimental data (filled symbol – adsorption; empty
symbol – desorption) and the lines represent the Sips model isotherm fitting. ........................... 60
Figure 4.28 - Selectivity of CO2/N2 as a function of pressure at 303.15K. .................................. 61
Figure 4.29 - Single-component adsorption isotherms for N2 (blue line), CO2 (green line), C4H10
(red line) on ANGUARD 6 and CO2 (purple line) on ACHM 6 at 303.15 K. ................................ 62
Figure 4.30 - Single-component adsorption isotherms for of CO2 on ANGUARD 6 and CO2 on
ACHM 6 at 303.15 K. The amount adsorbed is represented in moles of carbon dioxide by mass
of carbon sample. ........................................................................................................................ 63
Figure 4.31 - Single-component adsorption isotherms for of CO2 on ANGUARD 6 and CO2 on
ACHM 6 at 303.15 K. The amount adsorbed is represented in moles of carbon dioxide by
volume of carbon sample. ........................................................................................................... 64
Figure A.1- BET Surface Area Report and BET Surface Area Plot for ANGUARD 6, obtained
from DataMasterTM
, V4.00 (2004). The value of is indicated as Qm. .................................. 80
Figure A.2 - t-Plot Report and t-Plot for ANGUARD 6, obtained from DataMasterTM
, V4.00
(2004). ......................................................................................................................................... 80
Figure A.3 - Horvath-Kawazoe Report for ANGUARD 6, obtained from DataMasterTM
, V4.00
(2004). ......................................................................................................................................... 81
Figure A.4 - Density Functional Theory results for ANGUARD 6, obtained from DataMasterTM
,
V4.00 (2004). ............................................................................................................................... 81
Figure A.5 - Intrusion Data from Hg porosimetry for ANGUARD 6 ............................................. 81
Figure A.6 - Illustration of a piece of the ACHM used for the determination of its bulk density. . 83
Figure B.1 - Net (◊), excess (□) and total (∆) adsorption isotherms of nitrogen on ANGUARD 6
at 303.15K (top), 323.15K (middle) and 353.15K (bottom). ........................................................ 87
Figure B.2 -Net (◊), excess (□) and total (∆) adsorption isotherms of butane on ANGUARD 6 at
303.15K (top), 323.15K (middle) and 353.15K (bottom). ............................................................ 89
Figure B.3 - Net (◊), excess (□) and total (∆) adsorption isotherms of carbon dioxide on
ANGUARD 6 at 303.15K (top), 323.15K (middle) and 353.15K (bottom). .................................. 91
Figure B.4 - Net (◊), excess (□) and total (∆) adsorption isotherms of carbon dioxide on the
ACHM at 303.15K (top), 323.15K (middle) and 353.15K (bottom). ............................................ 93
Figure C.1 - Magnetic Suspension Balance (Metal version). ...................................................... 99
Figure C.2 - Pressure Transducers from MKS Baratron and Omegadyne. ................................ 99
Figure C.3 - Unit controller for data acquisition by Rubotherm GmBH. ...................................... 99
Figure C.4 - Gas Bottles from Air Liquid and Praxair. ................................................................. 99
Figure C.5 - Pressure Generator from HiP. ............................................................................... 100
Figure C.6 - Thermostatic Bath, Refrigerator/Heater from Julabo. ........................................... 100
Figure C.7 - Vaccum Pump from Edwards. ............................................................................... 100
Figure C.8 - Heater from Nabertherm. ...................................................................................... 100
Figure C.9 - Pictures of the experimental apparatus used in the equilibrium measurements. . 100
xviii
xix
List of Tables
Table 2.1 - Typical characteristics of activated carbons [36]. ..................................................... 16
Table 3.1 - PZC results obtained for ANGUARD 6. .................................................................... 19
Table 3.2 - PZC results obtained for the ACHM. ......................................................................... 19
Table 3.3 - Values of nCFS for RUN A, RUN B, RUN C and RUN D. ........................................ 23
Table 3.4 - (Continued) Values of nCFS for RUN A, RUN B, RUN C and RUN D. .................... 24
Table 3.5 - Results obtained from t-plot method for ANGUARD 6. ............................................. 29
Table 3.6 - Results obtained by HK method for ANGUARD 6. ................................................... 30
Table 3.7 - Results obtained from Mercury Porosimetry for ANGUARD 6. ................................ 33
Table 4.1 - Blank calibration of the measuring cells.................................................................... 44
Table 4.2 - Results obtained from helium measurements for ANGUARD 6 and ACHM. ........... 45
Table 4.3 - Parameters obtained from Sips isotherm models for the pure gases in ANGUARD 6
and ACHM. .................................................................................................................................. 55
Table A.1 - Results from Bohem Titrations Experiments for RUN A – 48 hours without
centrifugation (ANGUARD 6). ..................................................................................................... 77
Table A.2 - Results from Bohem Titrations Experiments for RUN B – 48 hours with
centrifugation (ANGUARD 6). ..................................................................................................... 77
Table A.3 - Results from Bohem Titrations Experiments for RUN C– 24 hours with centrifugation
(ANGUARD 6). ............................................................................................................................ 78
Table A.4 - Results from Bohem Titrations Experiments for RUN D – 48 hours with
centrifugation (ACHM). ................................................................................................................ 78
Table A.5 - N2 adsorption isotherm at 77K for ANGUARD 6. ..................................................... 79
Table A.6 - Characterization physical parameters of ANGUARD 6. ........................................... 82
Table B.1 - Experimental data obtained from helium measurements at 353.15K for ANGUARD 6
and ACHM. .................................................................................................................................. 85
Table B.2 – Experimental nitrogen adsorption equilibrium data on the carbon sample
ANGUARD 6 at 303.15K, 323.15K and 353.15K. 54 experimental data points were measured.
..................................................................................................................................................... 86
Table B.3 - Experimental butane adsorption equilibrium data on the carbon sample ANGUARD
6 at 303.15K, 323.15K and 353.15K. 40 experimental data points were measured. .................. 88
Table B.4 - Experimental carbon dioxide adsorption equilibrium data on the carbon sample
ANGUARD 6 at 303.15K, 323.15K and 353.15K. 43 experimental data points were measured.
..................................................................................................................................................... 90
Table B.5 - Experimental carbon dioxide adsorption equilibrium data on ACHM at 303.15K,
323.15K and 353.15K. 41 experimental data points were measured. ........................................ 92
Table C.1 - Characteristic of the several pressure transducers used in this work. ..................... 97
xx
xxi
List of Symbols
Notation
- Dispersion constants for the adsorbate calculated from Kirkwood-Muller formulae (erg2.cm
6)
- Dispersion constants for the adsorbent calculated from Kirkwood-Muller formulae (erg2.cm
6)
- Affinity constant (bar-1
)
- Affinity constant at a reference temperature, (bar-1
)
– BET Constant
- Gas molecular diameter ( )
- Adsorbent atom diameter ( )
- Average of the adsorbate and adsorbent molecule diameters ( )
- Distance between the nuclei of two parallel infinite lattice planes or pore width ( )
- Weight read from the balance at any time (g)
- Mass of the sample holder (g)
- Mass of the sample adsorbent (g)
- Kinetic energy of electron (0.8183x10-6
erg)
– Sips isotherm model parameter
- Sips isotherm model parameter at the same reference temperature,
- Amount adsorbed (cm3/g)
- Monolayer capacity (cm
3/g)
- Number of moles of carbon surface functionalities at the carbon surface
- Number of gas molecules per unit area (molecules/cm2)
- Avogadro’s number (6.023x1023
molecules-1
)
- Number adsorbent molecules per unit area (molecules/ cm2)
– Absolute pressure (bar or mmHg)
- Saturated vapor pressure (bar or mmHg)
- Adsorbed quantity of the more adsorbed specie (mol/kg)
- Quantity of the less adsorb quantity (mol/kg)
- Specific excess adsorbed (mol/kg)
xxii
- Net adsorption (mol/kg)
- Total amount adsorbed (mol/kg)
- Maximum amount adsorbed (mol/kg)
– Heat of adsorption (kJ/mol)
or - Ideal gas constant (8.31441x107 ergs/mole.K)
– BET surface area (m2/g)
– Surface external area (m2/g)
- Temperature (K)
- Standard multilayer thickness on the reference non-porous material at the corresponding
relative pressure ( )
Bulk volume (mL/g)
- Volume of all moving parts present in the measuring cell (cm3/g)
– Microporous volume (cm3/g)
- Accessible pore volume of the adsorbent (cm3/g)
- Total porosity (%)
Skeletal volume in Hg porosimetry (mL/g)
- Specific adsorbent volume impenetrable to the adsorbate (cm3/g)
- Diamagnetic susceptibility of an adsorbate atom (cm3)
- Diamagnetic susceptibility of gas molecule (cm3)
- Isosteric heat of adsorption (kJ/mol)
Greek Letters
α - Sips isotherm model parameter
- Polarizability of gas molecule (cm3)
- Polarizability of adsorbent atoms (cm3)
- Selectivity
- Cross-sectional area that each adsorbate molecule occupy in the completed monolayer
(nm2)
- Density of the sample holder (g/cm3)
- Skeletal density of the sample adsorbent (g/cm3)
xxiii
- Fractional loading
xxiv
1
Chapter 1
1. Introduction
1.1. Motivation
Carbon dioxide (CO2) is naturally present in the atmosphere as part of the Earth's
carbon cycle (the natural circulation of carbon among the atmosphere and life beings). But since
the Industrial Revolution, human activities have been altering the carbon cycle, by adding more
CO2 to the atmosphere. It is now estimated that around 90% of the carbon dioxide present in
the atmosphere is from anthropogenic origin [1].
According to the Inventory of U.S Greenhouse Gas Emissions and Sinks (1990-2011)
electricity production (38%), transportation of people and goods (31%) and industry (14%) are
the sectors that most contribute to the emission of carbon dioxide to the atmosphere. This
results from the burning of fossil fuels like coal, oil and natural gas [2]. Since carbon dioxide is a
greenhouse gas, which can contribute to global climate change, it is imperial to reduce its
emissions. In the past few decades many projects and possible solutions have been proposed
to mitigate this problem. This includes improvements in energy efficiency and utilization of
renewable and greener sources of energy [3].
Nowadays, CCS (CO2 Capture and Storage) is starting in several power plants [4]. CCS
process consists in the capture of the carbon dioxide resulting from the burning of fossil fuels.
CCS can be performed using pre-combustion or post-combustion techniques [5]. Then, after
being separated from other gases, CO2 is compressed and transported through a net of
pipelines or ships so it can be injected in underground geological formations, where it will be
safely storage for several years. It is estimated that CCS process can reduce the emissions of
carbon dioxide by 90% [6].
The leading technology for implementation of post-combustion CO2 capture is an
absorption-based capture process where the flue gas resulting from the burning of fossil fuels
(mainly constituted by carbon dioxide, nitrogen and water vapor) is fed to reactor that contains a
solvent, usually amines. This process is called amine scrubbing [7]. The solvent will react with
the CO2 and then is pumped to another tank where carbon dioxide and the solvent will be
separated (amine regeneration), so the solvent can be recycled to the first tank and the gas
2
compressed. Despite being a mature process, amine scrubbing presents some drawbacks.
Amine regeneration is very energy intensive, corrosion problems as well as the emission of
carcinogenic compounds. Therefore, alternative processes are needed to overcome these
difficulties [8].
Adsorption-based separation processes are present important alternatives for CO2
capture from flue gases. Among this processes, Pressure Swing adsorption (PSA) is an
important option. In this process the pressurized flue gas stream is passed through a porous
solid that preferentially adsorbs CO2. After this, by decreasing the pressure, CO2 is desorbed
and ready to be compressed [9].
The PSA performance and the power consumption of PSA is highly related with the
adsorbents used. This is why the development of adsorbents with high adsorption capacities,
high selectivity and good regenerability for CO2 adsorption/desorption is so important in the
design of the adsorption process. Many adsorbents have been studied for PSA application,
including zeolites, activated carbons and, more recently metal organic frameworks. Among
these materials activated carbons combine the advantages of being robust and also
unexpansive materials [10].
1.2. Thesis Structure
This thesis is divided in five chapters:
Chapter 1: Introduction
The content of this chapter intends to advertise the reader about the problems related with the
growing emissions of carbon dioxide to the atmosphere and how adsorption can be a viable
alternative solution to help solve this problem.
This chapter also summarizes the organization of this work.
Chapter 2: Background
This chapter begins with a small review of adsorption-related concepts and the techniques
traditionally employed in adsorption equilibrium studies. Finally, some of the adsorbent
materials available in the market are described. Since activated carbons were the materials
employed in this work, a more detailed section is devoted to this type of materials.
3
Chapter 3: Adsorbent Characterization
This chapter briefly explains the characterization methods used (Point of Zero Charge – PZC -
method, Bohem titrations, N2 adsorption at 77K (BET Surface Area Method, t-Plot Method,
Horvath-Kawazoe Method and Density Functional Theory Method) and Mercury Porosimetry
and summarizes the results obtained for the activated carbons studied.
Chapter 4: Adsorption Equilibrium
In this chapter adsorption equilibrium data for carbon dioxide, nitrogen and butane on activated
carbons are presented.
The apparatus and experimental procedure employed are described. Then, an explanation is
given about the equations behind the several amounts adsorbed considered. The concepts of
absolute, excess and net amount adsorbed are discussed and the corresponding results
obtained are presented. The experimental data obtained was fitted with the Sips isotherm model
and the obtained results are presented and discussed.
Chapter 5: Conclusions and Suggestions for Future Work
In this chapter all the conclusions obtained along this study are presented and some
suggestions are left to the future.
4
5
Chapter 2
2. Background
2.1. Adsorption
2.1.1. Definition
Adsorption can be defined as the process where some atoms, ions or molecules
present in a given fluid, gas or a liquid (adsorptive), adhere to the surface of a solid material
(adsorbent). Due the increase of the adsorptive compound concentration, the solid material will
be enriched with the molecules of the fluid phase. This molecules adsorbed on the surface of
the solid material can be referred as adsorbate.
The reverse process, called desorption, can be defined as the removal of adsorbate
from the adsorbent. Desorption can be promoted by the decreasing the pressure and/or
increasing the system temperature [11].
2.1.2. Some Applications
Adsorption phenomena is related with important technology processes used nowadays,
not only because some adsorbents are used in large scale as desiccants, catalysts or catalyst
supports but also because adsorption can be used in areas so diverse like separation of gases,
purification of liquids and pollution control, like the removal of aqueous contaminants from
groundwater [12]. This phenomenon is also useful for the determination of the surface area and
pore size distribution of a diverse range of powders and porous materials [13].
2.1.3. Chemisorption and Physisorption
Adsorption can be defined according to the nature of the interactions
adsorbate/adsorbent. Adsorption can be divided in two types: Chemisorption and Physisorption.
Chemisorption consists in the adsorption of molecules of a fluid on the surface of a solid caused
by covalent or ionic bonding. In physisorption the fluid molecules are retained due to Van der
Waal forces (including dipole–dipole, dipole-induced dipole and London forces). Due to their
different nature, physisorption and chemisorption can be distinguished by [11], [13]:
6
a) Since in chemisorption the molecules of the fluid phase react with the adsorbent
molecules, its original form is not kept. On the other hand, in physisorption, the
molecules are adsorbed and desorbed without any chemical reaction.
b) Physisorption has a relatively low degree of specificity and for that adsorbate molecules
can be linked to other adsorbate molecules and to the adsorbent, forming multilayers.
On the other hand, chemisorption is dependent on the reactivity of the adsorbent and
adsorptive, so adsorbate molecules can only linked to specific sites of the adsorbent
surface, being confined to a monolayer.
c) The energy of chemisorption has the same order of magnitude as the energy change in
a comparable chemical reaction. Physisorption is always exothermic, but the energy
involved is, generally, not much higher than the energy of condensation of the
adsorptive.
2.1.4. Adsorption Isotherms
The relation, at constant temperature, between the amount adsorbed and the
equilibrium pressure (for gases) is known by adsorption isotherm. When the adsorptive pressure
stabilizes, the equilibrium is reached, which means that the quantity of molecules adsorbed and
the molecules in the fluid phase will not vary with time
In order to describe this relation there are several models of isotherms. However the
isotherm models of Freudlich, Langmuir and BET (Brunauer, Emmett and Teller) are the most
commonly observed [14]. A typical gas adsorption isotherm is represented by a plot of the
amount adsorbed versus the adsorptive pressure. The pressure can also be expressed as a
ratio of the adsorptive pressure, P, to the saturated vapor pressure, .
Despite the multitude of different gas-solid systems available, the majority of the
isotherms obtained can be conveniently grouped into six classes according to the International
Union of Pure and Applied Chemistry (IUPAC) classification which is based on the original
Brunauer, Deming, Deming and Teller (BDDT) classification (1940) [13] . The different types
can be seen in Figure 2.1.
7
Figure 2.1 - IUPAC gas physisorption isotherm classification [13].
The different adsorption isotherm types are related with the adsorbent and adsorbate
properties. The differences between them can be listed as follow.
Type I - Observed in the physical adsorption of gases on microporous solids, in which
the pore size is not much greater that the molecular diameter of the adsorbate molecule. This
type of adsorption isotherm is common in activated carbons and black carbons [15].
Type II – Typically observed in non-porous or macroporous adsorbents. An inflexion
point, or knee, is indicated by point B in Figure 2.1. This point indicates the stage at which the
monolayer coverage is complete and multilayer adsorption begins to occur [16].
Type III – Observed in macroporous solids. This isotherm is convex to the ( / ) axis
over its entire range and therefore does not exhibit a point B. This feature is indicative of weak
adsorbent/adsorbate interactions. These kinds of isotherms are not common [13], [16].
Type IV – Characteristic of mesoporous adsorbents, this kind of adsorption isotherm
possess a hysteresis loop (which means that the adsorption isotherm is different from the
desorption isotherm). Type IV isotherms are common but the exact shape of the hysteresis loop
varies with the system properties [13], [15].
Type V – Like the previous type, this isotherm is observed mesoporous solids and
possesses a hysteresis loop. As Type III isotherm this kind of isotherm indicatives weak
adsorbent/adsorbate interactions. This type of isotherms is relatively rare [13], [15].
8
Type VI – Usually observed on porous solids with uniform surfaces, this kind of
isotherm is relatively rare and is associated with layer-by-layer adsorption on a highly uniform
surface [13].
This type of isotherm classification is only applicable to the adsorption of a single-
component gas within its condensable range of temperature. Such measurements are
extremely useful for the characterization of porous materials [13].
2.1.5. Measuring Adsorption Isotherms
In order to determine the adsorption isotherms and the energies associated with the
adsorption phenomena, experimental measurements must be made. Depending on the gas-
solid system in study and the operational conditions there are several gas adsorption methods
to quantify the amount adsorbed. The most used are the volumetric and gravimetric methods.
2.1.5.1. The Volumetric Method
The name volumetric method dates from the Emmett and Brunauer (1937) [13]
experiments which were made using a mercury burette and a manometer. This technique is
based on the measurement of the gas pressure in a calibrated constant volume at a known
temperature.
A typical volumetric apparatus, shown in Figure 2.2, possess two different chambers:
one for the adsorbent sample is placed and other for the calibrated charge volume. Initially, the
adsorbent contained in the adsorption cell (or chamber) is activated under the appropriate
conditions in order to remove the previously adsorbed species. Both the column and the
reservoir are maintained at the desired temperature. After this, the reservoir is charged with the
gas to a predetermined pressure. The valve between the reservoir and the column is then
opened, and the adsorption equilibrium is established between the solid and the gas; the final
equilibrium pressure is recorded [17].
One big advantage of the volumetric technique is that the apparatus is less costly than
the gravimetric method. Volumetric units only require high precision pressure transducers and
high precision volume measurements. On the other hand, modern gravimetric units provide
more precise measurements than the ones obtained by volumetric method.
9
Figure 2.2 - Generic volumetric apparatus [17].
2.1.5.2. The Gravimetric Method
The determination of the amount adsorbed by the gravimetric method using a spring
balance was first use by McBain and Bakr in 1926. The apparatus consisted in an adsorbent
bucket attached to the lower end of a fused silica spring, which was suspended within a vertical
glass tube. Nowadays, spring balances have been substituted by suspension magnetic
balances (MSB) [13].
The process is initiated by placing the adsorbent sample inside a basket. The material is
then activated preferentially in-situ, in vacuum at a desired temperature. After the sample is
cleaned from impurities, the first measurement will give weight of the pair basket + clean
adsorbent sample. After this, adsorbate is fed to the adsorption chamber and the sample is then
allowed to equilibrate at the desired pressure and temperature (at a gas molar density). The
signal from the microbalance is recorded under equilibrium conditions (pressure and
temperature are constant). The change in the microbalance signal is a result of adsorption
occurring on the solid surface and the total buoyancy force [17]. A generic gravimetric
apparatus is shown in Figure 2.3.
The gravimetric technique has some advantages and disadvantages. The primary
disadvantage is the cost; microbalances are very expensive. In spite of that, this equipment has
a high precision and accuracy which made it very useful for high-quality research work and pore
analysis. That is the reason why the gravimetric method was chosen for this experimental work.
10
Figure 2.3 - Generic gravimetric apparatus [17].
2.2. Adsorbates
The following gases were selected for the present work since they are adsorbates with
interest in typical adsorption applications. Moreover, those gases are of extreme importance in
the mitigation of the greenhouse gas emissions.
2.2.1. Carbon Dioxide
Carbon dioxide, CAS number [124-38-9], CO2, Mr 44.010 g/mol, with a boiling point of -
329.72K (56.57ºC) and a melting point of 193.33K (-78.92ºC) is a colorless, odorless, non-
flammable gas with a sour taste. At normal temperature, the carbon dioxide molecules are
relatively stable and do not readily break down into simpler compounds. However, the
substance is very sensitive to high temperatures, ultraviolet light, and electrical discharge [18].
Like it was referred in Chapter 1, the majority of the carbon dioxide present in the
atmosphere is a result of human activities as energy production, transportation of people and
goods and from the industrial processes [2]. Several measures and protocols, like the Kyoto
Protocol [19], to reduce the emissions of greenhouse gases that lead to global warming, are
being applied nowadays. Despite this fact, the emissions of CO2 to the atmosphere grow each
day. According to data from the Mauna Loa Observatory and the NOAA-ESRL (National
Oceanic & Atmospheric Administration, from U.S Department of Commerce), last year (data
from February 2013) the concentration of carbon dioxide present in the atmosphere was 396.88
ppm and this year, by the same time was 397.38 ppm instead of the 350 ppm, which is the goal
imposed since 1988 [20].
11
2.2.2. Nitrogen
When two molecules of elemental nitrogen, N (atomic number 7, Ar 14.0067 g/mol) form
a stable diatomic molecule they origin a molecular substance named nitrogen, CAS number
[7727-37-9], N2, Mr 28.0134 g/mol. At atmospheric pressure and room temperature, nitrogen is
a colorless, odorless, noncombustible gas. Nitrogen has a boiling point of 77K (at 1.01 bar) and
a melting point of 63.29K (−209.86 ºC).
Nitrogen, which means “lifeless” in Greek, was named by Lavoisier. This molecule is
obtained from air and is one of its major constituents (78%). In industry, cryogenic (low-
temperature) processes, adsorption processes (such as PSA – Pressure Swing Adsorption),
and membrane separation are used to separate nitrogen from air [18]. Also, N2 is one of the
main components of flue gases and, therefore, the knowledge of its adsorption properties is
extremely important for the modelling of adsorption-based processes for CO2 capture from flue
gases [1].
2.2.3. Butane
Butane, CAS number [106-97-8], C4H10, Mr 58.122 g/mol, with a boiling point of
273.65K (0.5ºC) is a gaseous hydrocarbon with a colorless and odorless aspect. This substance
currently known as n-butane (to indicate that the carbon atoms are linked in a straight chain)
occurs in natural gas and in crude oil. It is formed in large quantities, by catalytic cracking in the
refining of petroleum to produce gasoline. Commercially, n-butane can be added to gasoline to
increase its volatility [21]. Removal of butane from natural gas and biogas is extremely
important, reason why the study of its adsorption equilibrium is of major importance for the
design of adsorption based processes [22].
2.3. Adsorbents
2.3.1. General Adsorbents
Adsorbents are porous solid materials which have the ability to adsorb molecules from a
liquid or gas [14]. According to IUPAC [23], the pore size generally specified as pore width (the
available distance between two opposite walls) of a porous material can be classified as:
Micropore – Pore of internal width less than 2 nm;
Mesopore - Pore of internal width between 2 and 50 nm;
Macropore - Pore of internal width greater than 50 nm.
12
The potential of adsorbents has been studied since the 18th century [14] and the
applications for the use of adsorbents have grown with the years. For this reason, it became
necessary to design new adsorbents in order to face the specificity of the processes, which they
are applied. Some of the most important application for adsorbents are:
a) Gas separation processes like the upgrading of biogas [24];
b) Cleaning processes, sewage gas purification [25], removal of contaminants from
groundwater through adsorption [12] and the cleaning of industrial effluents [26];
c) Gas storage processes like Adsorbed Natural Gas (ANG) [22];
According to the gas adsorption process, a proper selection of the adsorbent must be
made. Nowadays there are several adsorbents available. Some of them are listed next.
The name activated alumina is generally applied to an alumina adsorbent prepared by
the heat treatment of some hydrated alumina (i.e. a crystalline hydroxide, oxide-hydroxide or
hydrous alumina gel) [13]. This material presents a good mechanical resistance and can be
used in moving bed applications [27]. The surface chemistry of activated alumina, as well as its
pore structure, can be modified by the use of a controlled thermal treatment [28].
Silica Gel has a granular and amorphous form. It is produced by heating a gel, product
of the acidification of a solution of sodium silicate. This glassy material is highly porous and it is
used to dry liquids and gases and also to recover hydrocarbons [27]. In addition, its surface can
be modified by reacting (or grafting) with a monomolecular layer of organic ligand. These
modified silica gels can be applied in several chromatographic applications [28].
Zeolites, also referred to as molecular sieves, are microporous crystalline solids with
well-defined structures. Generally they contain silicon, aluminum and oxygen in their framework
and cations, water and/or other molecules within their pores. Many zeolites occur naturally as
minerals as others are synthetic. The major use of zeolites are in petrochemical cracking, ion-
exchange (water softening and purification), and in the separation and removal of gases and
solvents [29].
Metal-Organic Frameworks (MOFs) are crystalline materials composed of two
components: metal ions or ion clusters and organic molecules known as linkers. The choice of
metal and linker has significant effects on the structure and properties of the particular MOF.
These materials have broad potential for industrial applications because of their attributes: large
surface-areas and the flexibility with which their structures can be varied [30]. These structures
can be used in studies for high-density storage of gases, including methane (natural gas) and
hydrogen. [31], [32], [33] .
13
Once the adsorbents used in this work are two activated carbons, a special section will be
dedicated to these materials.
2.3.2. Activated Carbon
2.3.2.1. Historical Aspects
The first known use for carbon dates from Egyptian time, where this material was used
for oil purification and medicinal purposes. By the early 19th century both wood and bone
charcoal were used in large-scale for the decolorization and purification of cane sugar [27], [28],
[34].
However, it was only in the beginning of the First World War (WWI) that the potential of
activated carbon was really capitalized upon. The advent of gas warfare necessitated the
development of suitable respiratory devices for personnel protection. Granular activated carbon
was used to this end as, indeed, it still is today [35]. By the late 1930’s there was considerable
industrial-scale use of carbon for gaseous and liquid phase application. During the Second
World War (WWII), a more sophisticated chemically impregnated carbon for entrapment of
nerve gases was produced [34].
2.3.2.2. Structure and Precursor Materials
Among all the adsorbents used in industry, activated carbon, also called activated
charcoal, is one of the most used. This microporous adsorbent can be obtained from several
carbon containing materials. Its high surface area (activated carbons can have BET-areas
larger than 2000 m2/g) [13] and its microporore volume, associated with the presence of variety
of functional groups on its surface, make activated carbon materials powerful adsorbent. The
structure of carbon it is basically comprised by graphitic plates, as showed in Figure 2.4. The
vertices and the edges can accommodate a range of elements such as oxygen, nitrogen and
hydrogen which comprises the surface functional groups. Its graphite structure is very important
from the adsorption capacity point of view, because it provides space on the slit-pore channels
to accommodate adsorbate molecules [36], [37].
14
Figure 2.4 - Hexagonal structure of graphite [13].
Depending on the raw materials used for its production, several types of activated
carbon can be obtained. Almost all materials containing high fixed carbon content can
potentially be activated. The most used carbonaceous source materials are coal (anthracite,
bituminous, sub-bituminous and lignite), coconut shell, peanut shell, wood, peat, coals,
petroleum coke, bones and fruit nuts. Among these, anthracite and bituminous coals have been
the major sources employed [27], [28], [38], [39].
2.3.2.3. Carbonization and Activation
The process for the production of activated carbon usually involves three steps: a) Raw
material preparation, b) Carbonization and c) Activation. In order to achieve the desired pore
structure and mechanical strength, the activation conditions must be carefully controlled. There
are two kinds of activation: physical activation and chemical activation. In both a step of
carbonization is required. This step allows the pure carbon to be extracted by pyrolysis [12],
[40].
In physical activation, once the material is carbonized it is exposed to oxidizing gases
like carbon dioxide, oxygen or steam, under a temperature usually between 1073.15K and
1273.15K. This activation step serves to create porosity allowing the tailoring of the desired size
distribution and surface area.
In chemical activation the material is first impregnated with chemicals agents such as
phosphoric acid or zinc chloride and then is carbonized [12], [28], [41].
During the manufacturing process, macropores are the first to be formed. This occurs
due the oxidation of weak points (edge groups) on the external surface area of the raw material.
Mesopores are then shaped and are, essentially, secondary channels formed in the walls of the
macropore structure. Finally, the micropores are molded by attack of the planes within the
structure of the raw material. All activated carbons contain micropores, mesopores, and
macropores (Figure 2.5) within their structures but the proportion vary according with the
precursor material and the activation conditions [39].
15
Figure 2.5 - Schematic representation of an activated carbon porous matrix [42].
2.3.2.4. Applications
Due to its unique adsorptive characteristics, activated carbon plays an important role in
many liquid and gas phase applications [43]. Some of the processes that use this adsorbent are
listed as following:
Because of its large surface area, purity and relative hardness, activated carbon is an
ideal carrier for catalytic metals, for example in batteries [44]. In the environmental field,
adsorption over activated carbon it is used for several applications. Some of these applications
are effluent treatment of industrial and municipal waste waters, air purification and capture of
volatile organic compounds (VOC’s) from diverse streams, and the removal of pesticides from
contaminated soils [12], [44]. In medicine, activated carbons can be employed in poisoning
treatments. Through its ingestion, this material prevents the poison from being absorbed in the
stomach. Sometimes, several doses of activated charcoal are needed to treat severe poisoning
[45].
Activated carbons present a great option for gas storage, especially for natural gas.
Because they have a large microporous volume, are efficiently compacted into a packed bed,
and can be cheaply manufactured in large quantities [22], [46]. This procedure permits storing
the gas at lower pressures, improving the safety criteria and reducing the compression costs
associated to the traditional storage methods [22].
Cane and sugar syrups require decolorization before being ready for final use.
Activated carbons are specially processed to develop pore structures that readily adsorb plant
pigments from the sugar (polyphenols) [47].
In summary, the characteristics of activated carbons make this type of materials one of
the best adsorbents that can be used in adsorption processes. Not only because of it high
surface areas and micropore volumes, but also because activated carbons can be produced in
several morphologies (beds, pellets, monoliths, fibers, etc. [8]). Activated carbons are also
16
available at low, prices, and being present in industry for so long, they are considered a robust
material for adsorption applications. In Table 2.1 it is possible to see some typical
characteristics of activated carbons.
Table 2.1 - Typical characteristics of activated carbons [36].
True density 2.2 g/cm3
Particle density 0.73 g/cm3
Total porosity 0.71 Macropore porosity 0.31 Micropore porosity 0.40 Macropore volume 0.47 cm
3/g
Micropore volume 0.44 cm3/g
Specific surface area 1200 m2/g
Mean macropore radius 800 nm Mean micropore half width 1-2 nm
In the next chapter, the adsorbents characteristics of the two activated carbons used in
this work will be properly study by using standard procedures.
17
Chapter 3
3. Adsorbent Characterization
3.1. Introduction
The design of a separation or purification process by adsorption begins with the choice
of a suitable adsorbent. The success or failure of the process is strictly related with the
performance of the adsorbent in both adsorption equilibria and kinetics. To satisfy these two
requirements the adsorbent must have [36]:
a) A reasonable high surface and a micropore volume, so it can have a good adsorption
capacity;
b) Relatively large pore network: If the pore size is too small the transport of the gas
molecules to the particle interior can take too long influencing the kinetics;
c) An easy desorption: If the adsorbent does not have properties that allow an easy
desorption, it will be necessary to expose the material to high temperatures or extremely
low pressure for its regeneration. This will contribute to reduce the adsorbent life due to
thermal ageing [15] and increases the energy consumption related to adsorbent
regeneration.
Therefore, in order to evaluate the adsorbent, adsorbent characterization must be
performed. Density, surface area, pore size distribution, pore volume, and the surface chemistry
are usually determined. In this study, an activated carbon, ANGUARD 6 (ANG 6), in the form of
extrudates with 1mm diameter, supplied by Sutcliffe Speakman Carbons Ltd. (UK) was
characterized at FCT/UNL using N2 adsorption at 77K, Mercury porosimetry, Bohem Titration
Method, Point of Zero Charge Method (PZC) and Thermogravimetric Analysis (TGA).
During the realization of this work, another carbonaceous material became available
and, therefore, was possible to perform the PZC and Bohem titrations analysis for this sample.
This material consists in an activated carbon honeycomb monolith (ACHM) purchased from
Master Carbon International Limited (UK). The monolith is cylindrical and presents 20 mm of
external diameter and 300 cells per square inch. Erro! Auto-referência de marcador inválida.
shows the two carbon samples.
Figure 3.1 - Activated carbons used in this study: ANGUARD 6 (on the left) and activated carbon honeycomb monolith (ACHM).
18
3.2. Characterization Methods
3.2.1. Point of Zero Charge (PZC)
3.2.1.1 General Description
The surface of a carbon particle is basic or acidic depending on the functional groups
that are in majority on its surface. When a small amount of well crushed activated carbon is
mixed with water, the ions H+ and OH
- from the dissociation of the functional groups are given to
the solution until the acid/base equilibrium is achieved. PZC (Point of Zero Charge) is a widely
used method to determine the surface nature of a carbonaceous adsorbent and can be defined
as the pH value at which a solid submerged in an electrolyte exhibits zero net electrical charge
[12], [48].
3.2.1.2. Experimental Procedure
Conditions:
An aqueous solution with 0.5 g of well crushed ANGUARD 6 and 50 mL of distillated
water was prepared. The glass container in which the solution was kept covered with an
aluminum sheet to prevent the oxidation of the carbon. After that, the solution was subject to
agitation during 48 hours, at 200 rpm. Thereafter, the agitation was stopped and the solution
allowed standing. Then, using a graduated pipette an aliquot was collected from the solution
and its pH value was determined using a digital pHmeter, CRISON 2001.
To secure the data reproducibility, three PZC experiments were made. In the first two,
the carbon was allowed to settle in the bottom of the vessel and the aliquots were collected from
the solution above the settled carbon. However, for the third experiment the solution was
subject to centrifugation and the supernatant removed. This procedure enhanced the time
needed to read pH in the digital pHmeter employed, since there were less carbon particles in
suspension.
PZC analysis was also performed for the activated carbon monolith. Since the amount
of material available for the experiments was less than for ANGUARD 6, the experimental
protocol had to be modified. This way, half the quantity of activated carbon and distilled water
were employed. Also, after the agitation step the carbon solutions were centrifuged.
.
19
3.2.1.3. Experimental Results and Data Analysis
The results obtained from PZC experiments for ANGUARD 6 and the ACHM are
presented in Table 3.1 and Table 3.2 respectively.
Table 3.1 - PZC results obtained for ANGUARD 6.
Experiment Weight of ANG 6 (g) pH value
1* 0.513 7.08 at 293.55K 2* 0.512 7.06 at 292.35K 3** 0.513 6.81 at 293.15K
*without centrifugation, ** with centrifugation, pH of distillated water: 5.35 at 293.55K.
Table 3.2 - PZC results obtained for the ACHM.
Experiment Weight of ANG 6 (g) pH value
1* 0.256 6.15 at 293.15K 2* 0.254 6.67 at 293.15K 3* 0.257 6.50 at 293.15K
* with centrifugation, pH of distillated water: 4.89 at 293.15K.
From the results obtained, the average pH value obtained for ANGUARD 6 was 6.98
and for the ACHM the average pH was 6.44. It can be concluded that the surface of both
activated carbons is amphoteric, which means that the acid and basic surface functional groups
are in equilibrium.
3.2.2. Bohem Titration Method
3.2.2.1 General Description
Many properties of carbon materials, in particular their adsorption behavior are
influenced by the chemisorbed oxygen. The oxygen present on the surface of an activated
carbon can bond with several elements such as oxygen, nitrogen, hydrogen and sulphur to form
functional groups. According to the activation method used, the functional groups present in the
surface of a carbon can be different. Acidic and basic surface sites usually coexist, but the
concentration of basic sites decreases with the increasing acid character of the surface and
vice-versa. The functional groups usually found in activated carbons are: carboxyl, carboxylic
anhydride, lactone, lactol, phenol, carbonyl (acidic groups) and chromene, ketone and pyrones
(basic groups) [15], [48] . Some of these functional groups can be seen in Figure 3.2 and
Figure 3.3.
20
Figure 3.2 - Simplified schematic of some acidic surface groups on an activated carbon [28].
Figure 3.3 - Schematic of some possible basic groups on an activated carbon [28].
The Boehm titration method permits the identification of the functional groups present in
the carbon surface. For this purpose, a small amount of activated carbon is mixed with some
strong bases in order to neutralize the phenols, lactonic groups and carboxylic acids. These
basic substances are NaOH, Na2CO3 and NaHCO3. Sodium hydroxide (NaOH) is the strongest
base and neutralizes all the Brönsted acids, while sodium carbonate (Na2CO3) neutralizes
carboxylic acids and lactonic groups (e.g. lactone) and sodium bicarbonate (NaHCO3)
neutralizes carboxylic acids. The number of basic sites is calculated from the amount of HCl
required to the titration [48].
3.2.2.2. Experimental Procedure
For determination of the acidic and basic surface functional groups of ANGUARD 6
three laboratorial experiments were performed. The experimental procedure was initiated by the
preparation of the basic and acidic solutions.
Solutions preparation:
Four solutions (three basic and one acidic) of 100 mL with concentration of 0.05 M were
prepared. For this purpose NaOH (Akzo Nobel, Eka Chemicals), Na2CO3 (Riedel-de-Häen,
99.5%), NaHCO3 (VR–V. Reis, Lda.) and HCl (Riedel-de-Häen, 37%) and distillated water (with
a pH of 5.35 at 293.55K) were used. The solutions were stored and used along the three
experiments in glass containers.
21
RUN A Procedure:
From each solution, 10 mL were extracted to glass containers and 1.0 g of well crushed
ANGUARD 6 was added. The glass containers were covered with an aluminum sheet to
prevent the oxidation of the carbon particles through the exposure to humid air. The solutions
were subjected to a period of agitation of 48 hours, at 200 rpm.
After this period, the agitation was stopped and the solutions were left to settle, during
10 to 15 minutes. In order to remove most of the carbon particles the solutions were decanted
several times. Then, aliquots were extracted with a graduated pipette and their pH values were
measured. Finally, the basic and acidic titrations were made, using as titrants an aqueous
solution of NaOH (0.1 M) and an aqueous solution of HCl (0.1 M). Two drops of phenolphthalein
were used as pH indicator.
RUN B Procedure:
The conditions and procedure used in this experiment were the same as the ones used
for RUN A, with the difference that after the agitation, the solutions were centrifuged to obtain a
better separation between the two phases (liquid and solid).
RUN C Procedure:
From each solution, 10 mL were extracted and 1.0 g of non-crushed ANGUARD 6 was
added. The glass containers were covered with aluminum sheet to prevent the oxidation of the
carbon particles through the exposure to air humidity. The solutions were subjected to a period
of agitation of 24 hours, at 200 rpm.
Thereafter, the agitation was stopped and the solutions were subject to centrifugation.
The aliquot was then extracted and the pH values were measured. Finally, the basic and acidic
titrations were made, using as titrants aqueous solutions of NaOH (0.1 M) and HCl (0.1 M).
Again, phenolphthalein was employed as pH indicator.
The calculations used for this analysis can be consulted in APPENDIX A.1.
3.2.2.3. Experimental Results and Data Analysis
The first attempt to perform these experiments was made using the Bohem titration
general procedure. Therefore, in the first experiment (RUN A), after the agitation was stopped,
the basic solutions with the carbon were left to settle for a period of time (10-15 minutes). Then
a volume from the supernatant was extracted to use in the pH measurement. This procedure is
22
inefficient because not only the waiting for the carbon mixture to settle adds substantial time to
the experiment but also the carbon particles left in the supernatant add difficulty the pH lecture.
The mass, volume and concentrations values obtained for the four experimental
procedures can be seen in APPENDIX A.2, Table A.1 to Table A.3 (for ANGUARD 6) and in
Table A.4 (for the ACHM). The values of for each functional group per gram of adsorbent
are presented in Table 3.3 and Table 3.4.
After the analysis of the results obtained for the four experiments, it was observed that
the values of for the lactones were always negative. Since there is no such thing as
negative number of moles, the values were considered to be zero. This means that the surface
of both activated carbons is not characterized by lactonic groups.
Along the four experiments it could be seen that the values for the quantity of basic
groups are always larger than the values for the quantities of carboxylic groups and phenols
isolated. But the overall quantity of the acidic groups is similar to the quantity of basic groups.
This is consistent with the PZC results, which showed that both ANGUARD 6 and the ACHM
have an amphoteric surface. In sum, the surface of both activated carbons is characterized by
basic groups (chromene, ketone and pyrones) and by acidic groups (carboxylic groups and
phenols, but not by lactonic groups).
23
Table 3.3 - Values of nCFS for RUN A, RUN B, RUN C and RUN D.
RUN A (ANGUARD 6)
nCFs Carboxilic Acids Lactones Phenols Total Basic Groups
(moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6)
NaOH 2.03E-04 2.61E-04 0 5.42E-05 3.61E-04
NaHCO3 2.61E-04
Na2CO3 1.49E-04 Total Acidic groups = 3.16E-04
HCl 3.61E-04 (moles/g of ANG6)
RUN B (ANGUARD 6)
nCFs Carboxilic Acids Lactones Phenols Total Basic Groups
(moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6)
NaOH 2.21E-04 3.64E-04 0 1.77E-04 4.43E-04
NaHCO3 3.64E-04
Na2CO3 4.35E-05 Total Acidic groups = 5.41E-04
HCl 4.43E-04 (moles/g of ANG6)
RUN C (ANGUARD 6)
nCFs Carboxilic Acids Lactones Phenols Total Basic Groups
(moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6) (moles/g of ANG6)
NaOH 1.90E-04 2.65E-04 0 1.83E-04 3.87E-04
NaHCO3 2.65E-04
Na2CO3 7.68E-06 Total Acidic groups = 4.48E-04
HCl 3.87E-04 (moles/g of ANG6)
24
Table 3.4 - (Continued) Values of nCFS for RUN A, RUN B, RUN C and RUN D.
RUN D (ACHM)
nCFs Carboxilic Acids Lactones Phenols Total Basic Groups
(moles/g of ACHM) (moles/g of ACHM) (moles/g of ACHM) (moles/g of ACHM) (moles/g of ACHM)
NaOH 1.17E-04 1.32E-04 0 8.45E-05 3.04E-04
NaHCO3 1.32E-04
Na2CO3 3.30E-05 Total Acidic groups = 2.17E-04
HCl 3.04E-04 (moles/g of ACHM)
25
3.2.3. Thermogravimetric Analysis (TGA)
3.2.3.1 General Description
Thermogravimetric analysis (TGA) is a procedure that allows the evaluation of the
physical and chemical properties of materials with the increase in temperature. Usually, the
weight of the analyzed sample is measured while the temperature is increased. The results are
generally plotted in a curve of the weight percentage versus temperature. Through this analysis
is possible to know the percentage of the impurities or volatile components, including humidity,
lost in the degassing process of an adsorbent. It is also possible to know which is the maximum
temperature to which an adsorbent can be subjected without contributing for its decomposition
[49]. This analysis can be performed in equipments that combine extremely precise balance and
a programmable furnace for temperature control.
3.2.3.2. Experimental Results and Data Analysis
A sample of ANGUARD 6 (8.4720 mg) was analyzed by TGA (TGA model Q50 V6.7
Build 203, Universal V4.4A TA Instruments - USA) to determine the temperature interval over
which the sample decomposes. This was done by recording the weight loss as a function of
increasing temperature. The analysis was performed under a nitrogen atmosphere at a heating
rate of 278.15K/min (5ºC/min).
The TGA profile obtained is showed Figure 3.4:
Figure 3.4 – TGA analysis of ANGUARD 6.
The TGA curve showed that the weight of ANGUARD 6 decreases steeply at 323.15K
(50ºC), and from about 323.15K to 823.15K (50ºC to 550ºC) the weight decreases slowly. After
26
873.15K (600ºC) the sample starts to decompose so it is clearly not advisable to heat the
adsorbent sample in the activation process more than 823.15K (550ºC). Employing adsorbent
activation temperatures between 323.15K and 473.15K (50ºC and 200ºC) the weight decrease
is around 3 to 4%.
3.2.4. Nitrogen adsorption at 77K
3.2.4.1 General Description
Nitrogen adsorption at 77K was performed for the activated carbon ANGUARD 6. The
experiment was performed using a static volumetric apparatus (ASAP 2010, Micromeritics
Adsorption Analyzer, USA) in a range of relative pressure 10-6
<P/P0<0.99. The sample weight
used in the experiment was 0.1418 g. The data from the isotherm was then analyzed using the
software DataMasterTM
, V4.00 (2004). The data obtained from the isotherm of N2 at 77K
measured are presented in APPENDIX A.3, Table A.5. Figure 3.5 shows the isotherm
obtained.
Figure 3.5 - Adsorption isotherm of N2 at 77K for ANGUARD 6.
3.2.4.1.1. BET Surface Area Method
Brunauer, Emmett and Teller (BET) method is commonly used to determine the surface
area of porous materials. Brunauer, Emmett and Teller (1938) extended the Langmuir
mechanism to multilayer adsorption and obtained an isotherm equation (BET equation). It is
assumed that the adsorbate molecules can settle on the adsorbent surface or on the top of
another adsorbate molecule [13].
0
100
200
300
400
500
600
700
800
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Am
ou
nt
adso
rbe
d (
cm3
/g)
Relative Pressure
ANG6 ads
ANG6 des
27
The first step is to determine the monolayer capacity, through the BET Equation
[13], [50]:
(Equation 3.1)
Where, is the amount adsorbed in cm3/g, is the absolute pressure in mmHg, is
the saturation pressure in mmHg and is the BET constant.
By simplifying Equation 3.1, a linear relation can be established between
and
.
(Equation 3.2)
This relation can be plotted, where the slope is s =
and the intercept is a =
.
By solving these two equations simultaneously, it can be obtained:
(Equation 3.3)
(Equation 3.4)
To guarantee the linear region of a BET plot it is recommended to restrict the values of
relative pressure to a range of 0.05-0.3 . However, the advisable procedure is to obtain, by
a statistical analysis, the best linear fit for the initial part of the isotherm [13].
Then, next step is to calculate the BET surface area, , the surface area that will
be available for adsorption, using that was obtained from Equation 3.3:
(Equation 3.5)
Where is the Avogadro’s number and is the cross-sectional area that each
adsorbate molecule occupy in the completed monolayer. For the nitrogen adsorption at 77 K the
value of is normally assumed to be 0.162 nm2. The value of is dependent on the nature
of the adsorbent-adsorbate and adsorbate-adsorbate interactions, the structure of the adsorbent
surface, and the operational temperature [13].
28
By restricting the relative pressure range between 0.025 to 0.31 , DataMasterTM
generated a BET Surface Area Report and a BET Surface Area Plot, shown in APPENDIX A.4,
Figure A.1.
The recommended procedure is to obtain the best linear fit for initial part of the
isotherm. Using only the experimental points in the BET relative pressure range, the fitting
obtained was not so good. By extending the relative pressure range, using the previous point
( = 0.025), the correlation coefficient obtained was higher (0.9976), indicating a better
fitting.
According to the BET theory, the BET constant is related exponentially to the enthalpy
(heat) of adsorption in the first adsorbed layer [51]. Because of this, the value must be
positive. If the value is negative this means that the relative pressure range chosen is not
adequate [13]. Since the value for obtained for this analysis was =106.50 (>0), the choice of
the relative pressure range was assumed to be adequate. Also, since the BET theory is an
extension of the Langmuir mechanism to multilayer adsorption, to consider the value
reliable, it is necessary that the knee of the isotherm is fairly sharp (i.e. the BET constant is
not less than ~100) (12).The BET surface area obtained for ANGUARD 6 is 1699.79 m²/g which
is within the typical range for activated carbons (between 300 and ∼ 4000 m2/g) [28].
3.2.4.1.2. t-plot Method
The t-plot method was proposed by Lippens and Boer in 1965. This method allows the
determination of micropore volume, external surface and micropore area. The experimental
from the isotherm is redrawn in a t-curve, i.e., a plot of the quantity of gas adsorbed as a
function of t, the standard multilayer thickness on the reference non-porous material at the
corresponding . These t-values are calculated using a thickness equation (Equation 3.6).
When the shape of the reference t-curve and experimental isotherm do not coincide, that is an
indication that a non-linear region was reached. From the point where the non-linear region
begins, a line is drawn (extrapolated) to intercept the yy axe (t=0). The values of external
surface area and micropore volume can be determined from the intercept and slope obtained
[13].
DataMasterTM
V4.00 uses Harkins and Jura Equation (1944) to determine the values of
thickness [52].
( ) √
(
)
(Equation 3.6)
29
Through the slope, s, of the extrapolation is possible to calculate the external surface
area in m2/g and from the intercept, a, the micropore volume in cm
3/g [16].
(Equation 3.7)
(Equation 3.8)
The micropore area, in m2/g can be calculated through the difference between the BET
surface area and the external surface area.
(Equation 3.9)
According to the literature [16] the t-plot method is valid for a relative pressure range of
0.08 to 0.75 . The results obtained for ANGUARD 6 can be seen in APPENDIX A.5,
Figure A.2. By restricting the relative pressures to the range recommended, the obtained
correlation coefficent was close to 1, which indicates a good fitting. According to the results
found in the literature [13], the values obtained are in acordance with the values for several
activated carbons. Table 3.5 presents a resume of micropore volume, external surface area and
micropore surface area for the activated carbon ANGUARD 6:
Table 3.5 - Results obtained from t-plot method for ANGUARD 6.
Micropore Volume
(cm3/g)
External surface area
(m2/g)
Micropore surface area
(m2/g)
0.94 52.50 1647.29
3.2.4.1.3. Horvath-Kawazoe (HK) Method
Horvath and Kawazoe (HK) described a semi-empirical, analytical method for the
determination of effective pore size distributions from N2 adsorption isotherms in microporous
materials. In its original form, the HK analysis was applied to nitrogen isotherms determined on
molecular sieve carbons, over the assumption that these adsorbents contained slit-shaped
graphitic pores. However, nowadays it can be applied to other adsorbents with different pore
geometries like zeolites [13], [16].
Horvath and Kawazoe found that the average potential could be related to the free
energy change of adsorption, creating a relation between filling pressure and the effective pore
width. Since the analysis is about an activated carbon, the following equation is based in the
slit-pore geometry [16]:
30
(
)
(Equation 3.10)
Much of the physical parameters present in this equation can be easily found in the
literature [28], [53], [54]. The nomenclature for equations Equation 3.10 to Equation 3.14 can
be found in List of Symbols (Page XXI to Page XXIII). Others like the dispersion constants and
the inter-nuclear distances must be calculated employing [16], [55]:
(Equation 3.11)
(Equation 3.12)
(
)
(Equation 3.13)
(
)
(Equation 3.14)
DataMasterTM
analysis allowed obtaining the results presented in Table 3.6:
Table 3.6 - Results obtained by HK method for ANGUARD 6.
Maximum Pore Volume
(cm3/g)
Medium Pore Diameter
0.98 18.4
The Horvath-Kawazoe detailed report for ANGUARD 6 is present in APPENDIX A.6.,
Figure A.3. The micropore size distribution can be seen in Figure 3.6.
31
Figure 3.6 - Micropore size distribution obtained from Horvath-Kawazoe Method for ANGUARD 6.
The pore size distribution results obtained from HK Method for ANGUARD 6 shows that
most of the pores lie in the micropore region (<20 ), but also reveals the existence of small
mesopores for pore widths slighty higher than 20 .
3.2.4.1.4. Density Functional Theory (DFT) Method
Density functional theory (DFT) is a quantum mechanical modelling method mostly used
in physics and chemistry areas. In chemistry, it has a great use for predicting a great variety of
molecular properties like molecular structures, vibration frequencies, atomization energies,
ionization energies, electric and magnetic properties, reactions paths, etc. [56] .
In adsorption science, this statistical method attempts to extend the accuracy of pore
size distribution analysis in both micropore and mesopore range [15]. DFT and Gibbs Ensemble
Monte Carlo molecular simulation (GEMC) represent an alternative to the classical methods,
like the HK method. Usually, for activated carbon it is assumed that the material is composed of
non-interconnected, slit-shapes pores with chemically homogeneous graphitic surfaces.
The pore size distribution obtained from DFT method is shown in Figure 3.7. The DFT
report for ANGUARD 6 can be seen in APPENDIX A.7, Figure A.4.
32
Figure 3.7 - Pore size distribution obtained from DFT Method for ANGUARD 6.
As it was seen for the HK Method, the DFT analysis confirms that the sample adsorbent
is mostly microporous, but the existence of mesopores with pore widths slightly above 20 is
confirmed.
3.2.5. Mercury Porosimetry
3.2.7.1 General Description
Mercury porosimetry characterizes the porosity of a given material by applying various
levels of pressure to a sample immersed in mercury. The pressure required to intrude mercury
into the sample pores is inversely proportional to the size of the pores. This indicates that at
first, macropores and mesopores are filled with mercury and just then, the mercury enters in the
micropores. In spite of the pressure applied, this method is reserved to the analysis of the
characteristics of larger pores, instead of the smaller pores [57].
3.2.7.2. Experimental Procedure
To perform this analysis, the sample is first loaded into a penetrometer and then, the
penetrometer is sealed and placed in a low pressure port, where the sample is evacuated to
remove air and moisture. The penetrometer’s cup is then automatically backfield with mercury.
As pressure increases, mercury intrudes into the sample’s pore, beginning with the pores that
have the large diameter. The instrument automatically collects low pressure measurements over
a range of pressures specified by the operator. Then, the penetrometer is moved to the high
pressure chamber, where high pressure measurements are taken [57].
33
3.2.7.3. Experimental Results and Data Analysis
A sample of ANGUARD 6 (0.1380 g) was subjected to mercury intrusion porosimetry,
using a mercury porosimetry penetrometer (model AutoPore IV 9500 V1.07) from Micromeritics
Instrument Corporation (USA).
The intrusion data summary from Hg porosimetry can be consulted in APPENDIX A.8,
Figure A.5. The intrusion-extrusion cycle can be seen in Figure 3.8.
Figure 3.8 - Experimental mercury intrusion-extrusion cycle for ANGUARD 6.
The curves give the volume of mercury (mL Hg/g of carbon sample) penetrated at a
given external pressure P into the measuring cell. The curve represented by the symbols
(−+−+−) indicates the intrusion curve and the other curve, represented by the symbols
(−⊖−⊖−) shows the extrusion curve. The results obtained in the mercury porosimetry analysis
are summarized in Table 3.7.
Table 3.7 - Results obtained from Mercury Porosimetry for ANGUARD 6.
Total Intrusion
Volume
(cm3/g)
Bulk Density
(g/ cm3)
Apparent Density
(g/ cm3)
Porosity
(%)
0.87 0.56 1.10 48.87
34
The total intrusion volume is the volume of mesopores and macropores (in case of
ANGUARD 6, the percentage of macropores it is very little, which can be seen in Figure 3.6
and Figure 3.7). Bulk density is the ratio between the mass of the sample to the sum of all the
volumes of the solid material (open, closed and blind pores), apparent density is the ratio
between the mass of the sample to sum of all the volumes, excluding open pores (closed and
blind). Figure 3.9 illustrates the definition of bulk and apparent densities. The percentage of
porosity is calculated through the Equation 3.15 [58].
(Equation 3.15)
Where is the total porosity, determined by difference between the bulk volume,
and the skeletal volume, , (which only considers the blind pores) [58].
Figure 3.9 - Illustration of bulk, apparent and skeletal densities [58].
Since the micropore volume was obtained through the HK method, it is now possible to
calculate the total pore volume for the carbon sample. By summing the micropore volume (0.98
cm3/g) and the total intrusion volume obtained by mercury porosimetry, the total pore volume for
ANGUARD 6 is 1.85 cm3/g.
3.3. Summary
With the purpose of summarizing the information obtained through the several methods
reported in this chapter (excluding PZC method, Bohem Titration method and TGA), APPENDIX
A.9, is presented.
35
Based on the information collected for the chemistry surface both activated carbons,
ANGUARD 6 and the ACHM, proved to have amphoteric surfaces, characterized by acidic
(except lactonic groups) and basic groups.
The thermogravimetric analysis showed that when heated until 323.15K (50ºC), the
weight of ANGUARD 6 decreases steeply, loosing at least 3% of its weight. Between 323.15K
to 823.15K (50ºC to 550ºC), the weight decreases slowly and after 873.15K (600ºC), the carbon
sample starts to decompose.
Through the BET Surface Area Method it was determined that ANGUARD 6 has a high
surface area, =1699.79 m²/g, a micropore volume of 0.98 cm3/g. The pore size
distribution, resulting from the Horvath-Kawazoe (HK) and Density Functional Theory (DFT)
methods, determined the existence of micropores, but also a small amount of mesopores with
pore widths slightly above 20 . The total pore volume of ANGUARD 6 was calculated by the
sum of the micropore volume (0.98 cm3/g) and the volume obtained from Mercury Porosimetry
(0.87 cm3/g) as a total of 1.85 cm
3/g. The high surface area and high micropore volume
obtained are good indicators that ANGUARD 6 will have a good adsorption capacity.
36
37
Chapter 4
4. Adsorption Equilibrium
4.1. Introduction
Adsorption equilibrium data is determinant information to understand an adsorption
process. No matter how many components are present in the system, the adsorption
equilibrium of pure components knowledge is essential to determine how much of those
components can be adsorbed on a solid adsorbent. This information can be used in the
modelling and optimization of adsorption separation processes [36].
4.2. Experimental Description
Adsorption equilibrium of carbon dioxide (CO2), nitrogen (N2) and butane (C4H10) at
303.15K, 323.15K and 353.15K and helium (He) pycnometry at 353.15K, in a range of 0 – 35
bar, were measured using an activated carbon, ANGUARD 6, as adsorbent. Additionally, a
structured activated carbon was purchased (an activated carbon honeycomb monolith) and,
therefore, adsorption data of carbon dioxide was also measured.
The adsorption equilibrium data were measured gravimetrically using a high-pressure
magnetic suspension balance (MSB) model ISOSORP 2000, from Rubotherm GmbH
(Germany), with automated online data acquisition in an in-house-developed Labview software
[59]. The advantage of the MSB is the possibility of accurately contactless weighing the
adsorbent samples, under nearly all environments. Instead of hanging the sample containing
baskets directly at the balance, the sample is coupled to a suspension magnet, achieving a
constant vertical position in a closed measuring cell. Using this freely suspension coupling, the
measuring force is transmitted contactless from the adsorption chamber to a Sartorius
microbalance, located outside, under ambient atmosphere.
Also, the MSB available in the group’s laboratory includes two baskets to allow the
measurement of adsorption equilibrium information for two adsorbents at the same time. The
device has a resolution of 0.01 mg, uncertainty lower than 0.002% of the measured value, and a
reproducibility of less than 0.03 mg to a maximum load of 25 g [60].
All gases employed were obtained from Air Liquide (Portugal) and Praxair (Portugal).
The purities of the gases employed are: CO2 > 99.998%, N2 > 99.995%, C4H10 > 99.95% and
He > 99.999%. Figure 4.1 shows the MSB and its components.
38
Figure 4.1 - Magnetic suspension balance components [60].
4.2.1. Adsorbent Sample Pre-Treatment
Prior to adsorption experiments, the adsorbents employed must be degassed (or
activated). This procedure ensures the removal of any adsorbed impurities and moisture.
The information obtained for the ANGUARD 6 carbon by thermogravimetric analysis
(TGA) showed that the activated carbon should be activated at temperatures between 323.15K
39
and 473.15K (50ºC and 200ºC). Within this range of temperatures the amount of loss weight is
3 to 4%. The TGA for the ACHM was not performed yet; reason why the temperature of
activation employed was the same as for the ANGUARD 6 carbon.
The ANGUARD 6 and ACHM samples were activated in situ at a temperature of
323.15K for a minimum of 4 hours, under vacuum. The heating up to 323.15K was performed at
a rate of 278.15K/min.
4.2.2. Experimental Apparatus
The measurements were carried out in the apparatus showed in Figure 4.2. The gas
enters the sealed chamber of the MSB and the pressure is registered at the exit line of the
apparatus, by several high accuracy sensors (PT), each one for a given pressure range. The
temperature is controlled using a double-jacket connected to a thermostatic bath (BATH). The
temperature is acquired using thermocouple (4-wire Pt100 probe) (T). A vacuum pump and a
set of valves (ball and check valves) are also coupled to the system in order to manage the
entrance and exit of gas, as well as the selection of the pressure transducer to use in each
measurement. The apparatus working range is limited to 150 bar and 373.15K. A complete
description of the equipment is given in APPENDIX C.
The adsorption laboratory apparatus is composed by four main units: a) Feed system
unit; b) Gravimetric unit with data acquisition; c) Pressure measuring and d) Temperature
measuring and control unit.
Feed Unit:
The feed unit is composed by a 1/8” OD SS tubing system prepared with a secondary
vacuum line and two feed lines, one for an inert, and another for the studied component.
The vacuum line is connected to a vacuum pump Edwards 5C. There is also a HiP pressure
generator to be applied whenever the desired pressure is higher than the available feed
pressure.
Gravimetric Unit:
The gravimetric unit is composed by the MSB, with acquisition of the weight values from
the microbalance, and simultaneous pressure data acquisition with a National Instruments PCI-
6023E Board. Acquisition is made using Labview construction, where the measurements are
monitored in order to see when equilibrium conditions have been reached.
40
Figure 4.2 - Schematic diagram of the experimental apparatus used in the equilibrium measurements.
Temperature Measuring and Control Unit:
The temperature control unit is composed by:
a) a refrigerator F32-HL from Julabo GmbH (Germany), keeping the temperature within 0.1
K of the set-point value;
c) a 4-wire Pt100 temperature probe, connected to the control unit of the MSB four-wire
Pt100 probes (RS Amidata, Spain), for temperature measurement on the measuring
cell.
Pressure Measuring and Control Unit:
The pressure was monitored by several pressure transducers (PT) of different ranges, granting good measurement accuracy:
1. MKS Baratron Type 627D Absolute Pressure Transducer, from MKS Baratron, used in
the pressure range of 2.00E-05 bar-1.32 bar;
2. PT PX01C1-150A5T pressure sensor (OM2), from Omegadyne Inc., used in the range
of 0-10 bar;
Pressure Generator
Vacum Pump
41
3. PT PX01C1-500A5T pressure sensor (OM3), from Omegadyne Inc., used in the range
of 0-35 bar;
4. PT PX03C1-3KA5T pressure sensor (OM4), from Omegadyne Inc., used in the range of
0-69 bar.
4.2.3. Experimental Procedure
The equilibrium adsorption data of carbon dioxide, CO2, nitrogen N2 and butane C4H10
and helium pycnometry was measured according to standard procedures:
a) Isothermal pressurization of the adsorption chamber of the MSB containing the carbon
sample with the pure gas up to the desired pressure, after previously preparation of the
system (3-4 g previously weighed several times using an analytical balance and an
average of these measurements was used to determine an initial mass of the sample;
this mass was confirmed at the initial start-up condition of the MSB);
b) With the in-house developed program [59], online measurement of pressure is taken.
Adsorption equilibrium is assumed to occur when, for a period of at least one hour, the
pressure, temperature, and sample weight do not vary. The measured weight is
recorded, and the amount adsorbed is determined. At this time, pressure and
temperature are also acquired and the value of the gas density (for this pressure and
temperature conditions) is obtained from a web database: NIST (National Institute of
Standards and Technology) [61];
c) The gas pressure in the adsorption chamber is then increased, and the sample is
allowed to reach equilibrium with this new condition. These measurements are repeated
until an entire adsorption isotherm is obtained;
d) After reaching the highest pressure point, the sample is degassed, once more step by
step, and the values of measured weight, temperature and pressure are taken, in order
to define the desorption isotherm;
e) In the end, after desorption isotherm is defined the sample is heated under vacuum to
ensure its complete regeneration.
42
4.3. Experimental Results and Data Analysis
The adsorption isotherms of carbon dioxide (CO2), nitrogen (N2) and butane (C4H10) at
303.15K, 323.15K and 353.15K and helium pycnometry at 353.15K, in a range of 0 – 35 bar
were measured on the ANGUARD 6 sample. Adsorption isotherms of CO2 in the same
temperature and pressure range were measured for the ACHM sample.
The gravimetric measurements, as well as all other conventional adsorption methods,
rather than giving the total amount adsorbed, , give the specific excess adsorbed, . The
is the total gas amount added into the measuring cell minus the amount that remains in the gas
phase upon system equilibration [30].
An alternative concept was developed by Gumma and Talu. This was named net
adsorption, and it is defined as the total amount of gas present in the measuring cell (with
the adsorbent), minus the amount that would be present in the empty cell (without the
adsorbent), at the same pressure and temperature (P, T) conditions. This parameter can be
calculated directly from the experimental data since it is independent of the adsorbent
characteristics such as pore volume, solid matrix density and impenetrable pore volume [17].
Therefore, this method eliminates the influence of the use of probe molecules in reporting
adsorption equilibrium data.
In order to determine the absolute adsorption, the buoyancy corrections must be
performed. Buoyancy corrections are considered to correct the influence of the gas density on
the measurements of the apparent weight sample. The corrections in the forces acting in the
sample holder, solid adsorbent and adsorbed phase are taken in account.
In the case of net adsorption, the buoyancy correction is needed only for the forces
acting on the sample holder, obtained through the blank experiments performed at different
pressures with the empty holder; for the excess adsorption, the buoyancy correction is
necessary for the impenetrable solid volume, which results in an apparent weight loss. This
correction is estimated as a product of the skeletal volume of the adsorbent and the gas density;
for the absolute adsorption the correction of the buoyancy acting on the pore volume is also
needed [17], [22] .
The weight, , reading from the balance at any time can be written as [22]:
(
) (
) (Equation 4.1)
With and representing the mass and the density of the sample holder, and
are respectively the mass and skeletal density of the sample adsorbent, is the density of the
43
bulk gas at equilibrium pressure and temperature. The specific excess adsorption, can be
calculated by:
(
) (Equation 4.2)
Here, is the volume of all moving parts present in the measuring cell (such as the
holding basket).
The blank experiments with the empty holder give its mass, volume and density from
the intercept and the slope of the linear decrease of apparent weight, , versus the gas density,
.
(Equation 4.3)
The values of and were estimated at 293.78K using He, helium. Figure 4.3 and
Figure 4.4 show the results obtained for the blank calibration of the adsorbent sample holders.
Table 4.1 summarizes the obtained results.
Figure 4.3 - Blank calibration of sample holder #1 used in the adsorption, using helium at 293.63K.
44
Figure 4.4- Blank calibration of sample holder #2 used in the adsorption, using helium at 293.63K.
Table 4.1 - Blank calibration of the measuring cells.
Sample holder Conditions (g) (cm3) (g/cm
3)
# 1 293.78 K. He 5.574 0.703 7.931
# 2 293.78 K. He 6.412 0.808 7.934
The determination of the mass and the density of the carbon sample ( and ) was
performed through high temperature (353.15K) measurements [22]. It is assumed that helium
acts as a probe molecule that penetrates into all accessible pore volume of the carbon without
being adsorbed.
(Equation 4.4)
Figure 4.5 shows the helium measurements performed for ANGUARD 6 and the
ACHM. The results obtained from the data analysis are listed in Table 4.2.
45
Figure 4.5 - Helium measurements on ANGUARD 6 (top) and ACHM (bottom) at 353.15K.
The values of weight and gas density, for both adsorbents are presented in APPENDIX B,
Table B.1.
Table 4.2 - Results obtained from helium measurements for ANGUARD 6 and ACHM.
Sample Conditions (g) (g/cm3)
ANGUARD 6 353.15 K, He 0.490 2.622
ACHM 353.15 K, He 0.424 2.847
The net adsorption can be calculated by (Equation 4.5. The excess amount is
related with through Equation 4.6 [17]:
(Equation 4.5)
y = -0.001x + 6.9018 R² = 0.9975
6.896
6.897
6.898
6.899
6.900
6.901
6.902
6.903
0 1 2 3 4 5
We
igh
t (g
)
ρg (kg/m3)
He at 353.15K on ANG 6
y = -0.0009x + 5.9986 R² = 0.9941
5.9945.9955.9955.9965.9965.9975.9975.9985.9985.9995.999
0 1 2 3 4 5
We
igh
t (g
)
ρg (kg/m3)
He at 353.15K on ACHM
46
(Equation 4.6)
Finally, the absolute amount adsorption (or total amount adsorbed), can be
calculated by the relation between and by [30]:
(Equation 4.7)
Where is the accessible pore volume of the adsorbent and
is the specific
adsorbent volume impenetrable to the adsorbate. For ANGUARD 6, is 0.98 cm3/g, calculated
in Chapter 3 by the Horvath-Kawazoe method.
During this study the of the ACHM was not determined. Therefore, a value
reported in the literature was employed. The used is 0.98 cm3/g. This is the value reported in
the literature for a similar ACHM also manufactured by Mast Carbon (UK) [10].
As an example, Figure 4.6 illustrates the net, excess and total amounts adsorbed for
nitrogen at 303.15K for ANGUARD 6. It is possible to see that in the low pressure region the
values do not shown significant difference. On the other hand, when the pressure increases the
values diverge from each other. For the high pressure region, the net adsorption is the lowest of
the three quantities, followed by excess adsorption and total adsorption. The same trend was
obtained for all the isotherms measured independently from the temperature and adsorbate
employed. All the data is presented in APPENDIX B, in, Figure B.1 to Figure B.4. Table B.2 to
Table B.5 shows the experimental values of , and for each pure gas.
Also it can be observed that as predicted in Chapter 3, the isotherms obtained are Type
I, which is characteristic of microporous adsorbents. Given that physical adsorption is an
exothermic phenomenon, the slope of the curvature of the adsorption isotherms will decrease
with the increasing of the temperature [22].
47
Figure 4.6 - Net (◊), excess (□) and total (∆) amount adsorbed for nitrogen (N2) at 303.15K for ANGUARD 6. The solid symbols represent adsorption and the open desorption.
4.3.1. Sips Isotherm Model
Knowledge of the adsorption equilibrium and heat of adsorption are essential for proper
design of any gas-phase adsorption process [22]. In fact, in adsorption separation processes,
the heat of adsorption is very important since during the process the heat is released and the
energy is partly absorbed by the solid adsorbent and partly is dissipated to the surrounding. The
portion absorbed by the solid increases the particle temperature, decreasing its local capacity
and broadening the mass transfer zone [36].
Several isotherm models are available to correlate experimentally obtained data. The
Sips isotherm model (or Langmuir-Freundlich model) is an extension of the Freundlich equation,
given by:
(Equation 4.8)
Where is the amount adsorbed in mole per unit mass or volume, is the maximum
amount adsorbed, is the affinity constant and measure how strong the adsorbate molecule is
attracted on to a surface, is the pressure, the parameter , usually greater than the unity,
characterizes the interaction between adsorbate/adsorbent and its magnitude increases with the
heterogeneity of the system.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0 5 10 15 20 25 30 35
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
N2 adsorption isotherm on ANG6 at 303.15K
qnet qex qt
48
The affinity constant and the parameter can be written in function of the temperature
as:
(Equation 4.9)
(Equation 4.10)
Here is the affinity constant at a reference temperature, ( , is the parameter at
the same reference temperature, is a constant parameter and is the ideal gas constant.
The isosteric heat of adsorption can be obtained by applying the Van’t Hoff’s equation,
which relates the adsorption heat effects to the temperature dependence of the adsorption
isotherm,
(Equation 4.11)
Where is the fractional loading (
. In terms of the pressure, isosteric heat it is
given by:
( ) (Equation 4.12)
And in terms of fractional loading,
( ) (
)
(Equation 4.13)
In the Sips isotherm model, parameter corresponds to the isosteric heat of adsorption
at the fractional loading of 0.5.
4.3.2. Experimental Results and Discussion
The experimental adsorption equilibrium data (the absolute amount adsorbed) was fitted
by the Sips adsorption isotherm model. The maximum amount adsorbed and the isosteric heat
of adsorption from each pure gas was determined. The Sips isotherm model is widely used to
describe data of many substances on activated carbon with good success [36].
49
To determine the Sips isotherm parameters for each adsorbate, the experimental
adsorption data was fitted using the software TableCurve 3D, v.4.0. This software combines a
powerful surface fitter that has the ability to describe three dimensional empirical data. Using
this software feature the absolute amount adsorbed (mol/kg) and the pressure (bar) are plotted
in order to the three working temperatures (K). From the fitted isotherm and the experimental
data it was possible to determine the average relative error percentage, (Equation
4.14), between the experimental points and the ones given by the isotherm model.
∑
(Equation 4.14)
The experimental and theoretical results obtained are disclosed in the following
sections.
4.3.2.1. Experimental data fitting employing Sips isotherm model
Nitrogen Adsorption on ANGUARD 6
Figure 4.7 - Sips model fitting of the N2 experimental data at 303K, 323K and 353K on ANGUARD 6 and parameters obtained. Symbols represent the experimental data and the surface is the global isotherm model.
50
Figure 4.8 - Single component N2 isotherms at 303K, 323K and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 6.98, 9.01 and 3.55, respectively. The N2 overall ARE error is 6.51%.
Figure 4.9 – Logarithmic representation of the single component N2 isotherms at 303K, 323K and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 6.98, 9.01 and 3.55, respectively. The N2 overall ARE error is 6.51%.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
Nitrogen fitting on ANGUARD 6
0.00
0.01
0.10
1.00
10.00
0.00 0.01 0.10 1.00 10.00 100.00
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
Nitrogen fitting on ANGUARD 6
51
Butane Adsorption on ANGUARD 6
Figure 4.10 - Sips model fitting of the C4H10 experimental data at 303K, 323K and 353K on ANGUARD 6 and parameters obtained. Symbols represent the experimental data and the surface is the global isotherm model.
Figure 4.11 - Single component C4H10 isotherms at 303K, 323K and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 3.90, 6.38 and 3.62, respectively. The C4H10 overall ARE error is 4.63%.
0
1
2
3
4
5
6
7
8
9
0 2 4 6 8 10
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
Butane fitting on ANGUARD 6
Parameter Value
qts (mol/kg) 17.117
b0 (bar-1) 0.447
Q (kJ/mol) 29.580
n0 2.409
α 0.336
T0 (K) 303.19
FSE 0.1426
r2 0.9971
Butane
52
Figure 4.12 - Logarithmic representation of the single component C4H10 isotherms at 303K, 323K and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 3.90, 6.38 and 3.62, respectively. The C4H10 overall ARE error is 4.63%.
Carbon Dioxide Adsorption on ANGUARD 6
Figure 4.13- Sips model fitting of the CO2 experimental data at 303K, 323K and 353K on ANGUARD 6 and parameters obtained. Symbols represent the experimental data and the surface is the global isotherm model.
0.10
1.00
10.00
0.00 0.01 0.10 1.00 10.00
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
Butane fitting on ANGUARD 6
53
Figure 4.14 - Single component CO2 isotherms at 303K, 323K and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 6.10, 6.20 and 8.45, respectively. The CO2 overall ARE error is 6.92%.
Figure 4.15 - Logarithmic representation of the single component CO2 isotherms at 303K, 323K and 353K on the activated carbon ANGUARD 6.Symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 6.10, 6.20 and 8.45, respectively. The CO2 overall ARE error is 6.92%.
0
2
4
6
8
10
12
14
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
Carbon dioxide fitting on ANGUARD 6
0.01
0.10
1.00
10.00
100.00
0.01 0.10 1.00 10.00 100.00
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
Carbon dioxide fitting on ANGUARD 6
54
Carbon Dioxide Adsorption on ACHM
Figure 4.16 - Sips model fitting of the CO2 experimental data at 303K, 323K and 353K on ACHM and parameters obtained. Symbols represent the experimental data and the surface is the global isotherm model.
Figure 4.17 - Single component CO2 isotherms at 303K, 323K and 353K on the activated carbon ACHM. Symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 5.29, 4.33 and 6.49, respectively. The CO2 overall ARE error is 5.37%.
0
2
4
6
8
10
12
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
Carbon dioxide fitting on ACHM
55
Figure 4.18 - Logarithmic representation of the single component CO2 isotherms at 303K, 323K and 353K on the activated carbon ACHM. Symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and lines represent the fittings with the Sips model. The %ARE errors for 303.15K, 323.15K, and 353.15K are 5.29, 4.33 and 6.49, respectively. The CO2 overall ARE error is 5.37%.
Through the several fittings presented between Figure 4.7 to Figure 4.18 it is possible
to see the good agreement between the fittings with the Sips isotherm model and the
experimental data through all the pressure range studied (including the low pressure region).
This information is corroborated by the obtained values of (the correlation coefficient), that
are close to 1 and the values the fit standard error ( ), which are very small. These
parameters of evaluation of the goodness of the fitting are showed in Table 4.3, along with the
fitting parameters for each adsorbate studied.
Table 4.3 - Parameters obtained from Sips isotherm models for the pure gases in ANGUARD 6 and ACHM.
ANGUARD 6 ACHM
Parameter
Nitrogen
Butane Carbon
Dioxide
Carbon Dioxide
qts (mol/kg) 11.96 17.12 21.66 11.80
b0 (bar-1
) 0.01 0.45 0.04 0.24
Q (kJ/mol) 9.02 29.58 19.80 22.85
n0 1.10 2.41 1.09 1.23
α 0.346 0.336 5.01E-05 1.68E-06
T0 (K) 303.21 303.19 303.22 303.22
FSE 0.0268 0.1426 0.1798 0.1001
r2 0.9996 0.9971 0.9979 0.9992
0.10
1.00
10.00
100.00
0.01 0.10 1.00 10.00 100.00
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
Carbon dioxide fitting on ACHM
56
4.3.2.2. Isosteric Heat of Adsorption employing Sips isotherm model
Nitrogen Isosteric Heat of Adsorption on ANGUARD 6
Figure 4.19 - Single-component isosteric heat of adsorption for N2 at 303.15K, 323.15K and 353.15K as a function of fractional loading on the activated carbon ANGUARD 6, predicted by Sips isotherm model.
Figure 4.20 - Single-component isosteric heat of adsorption for N2 at 303.15K, 323.15K and 353.15K as a function of equilibrium pressure on the activated carbon ANGUARD 6, predicted by Sips isotherm model.
8
10
12
14
16
18
20
0.00 0.10 0.20 0.30 0.40
(-∆
H)
(kJ/
mo
l)
Fractional loading, θ
Nitrogen on ANGUARD 6
303.13K
323.15K
353.15K
8
10
12
14
16
18
20
0 10 20 30 40
(-∆
H)
(kJ/
mo
l)
Pressure (bar)
Nitrogen on ANGUARD 6
303.13K
323.15K
353.15K
57
Butane Isosteric Heat of Adsorption on ANGUARD 6
Figure 4.21 - Single-component isosteric heat of adsorption for C4H10 at 303.15K, 323.15K and 353.15K as a function of fractional loading on the activated carbon ANGUARD 6, predicted by Sips isotherm model.
Figure 4.22 - Single-component isosteric heat of adsorption for C4H10 at 303.15K, 323.15K and 353.15K as a function of equilibrium pressure on the activated carbon ANGUARD 6, predicted by Sips isotherm model.
20
25
30
35
40
45
50
55
60
65
70
0.00 0.10 0.20 0.30 0.40 0.50 0.60
(-∆
H)
(kJ/
mo
l)
Fractional loading, θ
Butane on ANGUARD 6
303.15K
323.15K
353.15K
20
25
30
35
40
45
50
55
60
65
70
0 2 4 6 8 10
(-∆
H)
(kJ/
mo
l)
Pressure (bar)
Butane on ANGUARD 6
303.15K
323.15K
353.15K
58
Carbon Dioxide Isosteric Heat of Adsorption on ANGUARD 6
Figure 4.23 - Single-component isosteric heat of adsorption for CO2 at 303.15K, 323.15K and 353.15K as a function of fractional loading on the activated carbon ANGUARD 6, predicted by Sips isotherm model.
Figure 4.24 - Single-component isosteric heat of adsorption for CO2 at 303.15K, 323.15K and 353.15K as a function of equilibrium pressure on the activated carbon ANGUARD 6, predicted by Sips isotherm model.
19.800
19.800
19.800
19.800
19.801
19.801
19.801
19.801
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
(-∆
H)
(kJ/
mo
l)
Fractional loading, θ
Carbon Dioxide on ANGUARD 6
303.15K
323.15K
353.15K
19.800
19.800
19.800
19.800
19.801
19.801
19.801
19.801
0 10 20 30 40
(-∆
H)
(kJ/
mo
l)
Pressure (bar)
Carbon Dioxide on ANGUARD 6
303.15K
323.15K
353.15K
59
Carbon Dioxide Isosteric Heat of Adsorption on ACHM
Figure 4.25 - Single-component isosteric heat of adsorption for CO2 at 303.15K, 323.15K and 353.15K as a function of fractional loading on the activated carbon ACHM, predicted by Sips isotherm model.
Figure 4.26 - Single-component isosteric heat of adsorption for CO2 at 303.15K, 323.15K and 353.15K as a function of equilibrium pressure on the activated carbon ACHM, predicted by Sips isotherm model.
22.848
22.848
22.848
22.848
22.848
22.848
22.848
22.848
22.848
22.848
0.0 0.2 0.4 0.6 0.8 1.0
(-∆
H)
(kJ/
mo
l)
Fractional loading, θ
Carbon dioxide on ACHM
303.15K323.15K353.15K
22.848
22.848
22.848
22.848
22.848
22.848
22.848
22.848
22.848
22.848
0 10 20 30 40
(-∆
H)
(kJ/
mo
l)
Pressure (bar)
Carbon dioxide on ACHM
303.15K323.15K353.15K
60
From Figure 4.19 to Figure 4.26, it is possible to visualize the variation of the isosteric
heat of adsorption for each gas with the fractional loading and with pressure. For both cases,
the heat of adsorption decreases with the increasing of the fractional loading and with pressure.
This fact is related with the definition of isosteric heat. Activated carbons are characterized by
heterogeneous surfaces with a distribution of adsorption sites of different energies. Since the
adsorbate molecules begin to adhere to the surface sites with the highest energy, more energy
is released at lower surface coverage values. When those sites are totally fulfilled with
adsorbate, the gas molecules will be adsorbed on the remaining sites, with lower energy. So at
lower coverage, more energy is released reason why the isosteric heat of adsorption is higher
for lower amounts adsorbed and decreases with the enhancement of loading and also pressure
[22]. In spite of forming a plateau as the previous adsorbates, it can be seen in Figure 4.23 to
Figure 4.26 that of the values of isosteric heats of adsorption for carbon dioxide, in ANGUARD
6 and in ACHM almost not vary. This is due to the values of parameter obtained for the two
adsorbates (consult Table 4.3). Since the values are almost zero, the contribution of the
fractional loading and of the pressure is practical nil, reason why the values of isosteric heat of
adsorption almost not vary (see Equation 4.12 and Equation 4.13).
4.3.2.3. Data interpretation from Sips isotherm model
Starting by the maximum amount adsorbed, , from Table 4.3 it is possible to see that
for ANGUARD 6, CO2 has the highest saturation amount adsorbed, 21.66 mol/kg, followed by
C4H10, 17.12mol/kg and finally by N2 with 11.96 mol/kg. For comparison purposes the isotherms
of CO2, N2 and C4H10 on ANGUARD 6 at 303.15K are showed in in Figure 4.27.
Figure 4.27 - Single-component adsorption isotherms of N2, C4H10 and CO2 on ANGUARD 6 at 303.15 K. The symbols represent the experimental data (filled symbol – adsorption; empty symbol – desorption) and the lines represent the Sips model isotherm fitting.
0
2
4
6
8
10
12
14
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
61
The data shows that at low pressures, C4H10 is significantly more adsorbed than CO2
and N2. Despite this fact, the CO2 adsorption increases significantly with the pressure increase,
especially when compared with the capacity of the adsorbent towards N2. This indicates that
this activated carbon has potential to perform CO2/N2 separation. To evaluate this potential the
equilibrium selectivity of the activated carbon ANGUARD 6 for the CO2/N2 separation was
evaluated. Adsorption equilibrium selectivity results from a ratio between , the adsorbed
amount of the more adsorbed species and the adsorbed amount of the less adsorb quantity, .
(Equation 4.15)
Figure 4.28 shows the equilibrium selectivity for the CO2/N2 separation on ANGUARD
6. It can be observed that the selectivity for carbon dioxide decreases with the increasing of
pressure. The plot shows that the selectivity of CO2 over N2 is more favored at low pressures,
although it is known that the amount adsorbed increases significantly with pressure and both
these parameters must be considered when designing adsorption-based separation processes.
Figure 4.28 - Selectivity of CO2/N2 as a function of pressure at 303.15K.
The heats of adsorption, , predicted by the Sips isotherm model are 9.02 kJ/mol, 29.58
kJ/mol, 19.80 kJ/mol respectively for N2, C4H10 and CO2 dioxide on ANGUARD 6 and 22.85
kJ/mol for carbon dioxide on the ACHM. This means that the adsorbate molecules of butane are
the ones that release a higher amount of energy in the adsorption process.
For both activated carbons, for the four adsorbates the value of the Sips model
parameter is always greater than the unit. This parameter regards the system
heterogeneity. The larger is this parameter, the higher is the degree of heterogeneity. However,
2
2.5
3
3.5
4
4.5
0 5 10 15 20 25 30 35
Sele
ctiv
ity
Pressure (bar)
CO2/N2
62
this information does not point to what is the source of the heterogeneity, whether it be the solid
structural property, the solid energetically properties or the adsorbate properties [36]. The
higher value of parameter was obtained for butane on ANGUARD 6 ( = 2.41).
This same parameter influences the curvature of the adsorption isotherm. According
to the literature [36], the larger is the value of , the more nonlinear is the adsorption isotherm.
When the parameter is getting larger than 10, the adsorption isotherm is approaching a so-
called rectangular isotherm (or irreversible isotherm). The term "irreversible isotherm" is
normally used because the pressure needs to be decreased to an extremely low value before
adsorbate molecules would desorb from the surface. Such dramatic pressure decrease is quite
energy intensive and will therefore affect the energy consumption and cost of a separation
involving adsorbate-adsorbents with such isotherms. Figure 4.29 shows the adsorption
isotherm for each adsorbate and its corresponding values of parameter. It is possible to
visualize that butane has the squarest adsorption isotherm.
Figure 4.29 - Single-component adsorption isotherms for N2 (blue line), CO2 (green line), C4H10 (red line) on ANGUARD 6 and CO2 (purple line) on ACHM 6 at 303.15 K.
Through Table 4.3, it is also possible to see that butane has the highest value of , the
affinity constant for a reference temperature on ANGUARD 6. This means that the molecules of
butane are the ones that are more attracted to the surface of this activated carbon. Butane is
followed by carbon dioxide and nitrogen. Making a comparison between with the value of
obtained for CO2 on the ACHM, it can be concluded that the molecules of this adsorbate (CO2)
have more affinity for to the surface of the monolith than for the surface of ANGUARD 6.
0
2
4
6
8
10
12
14
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
𝑛 = 2.41
𝑛 = 1.09
𝑛 = 1.23
𝑛 = 1.10
63
Figure 4.30 - Single-component adsorption isotherms for of CO2 on ANGUARD 6 and CO2 on ACHM 6 at 303.15 K. The amount adsorbed is represented in moles of carbon dioxide by mass of carbon sample.
Figure 4.30 presents the CO2 isotherms at 303.15K of the two activated carbon
samples studied. It can be clearly observed that the ACHM adsorbs more CO2 at low pressures
while for higher pressures the ANGUARD 6 activated carbon presents higher adsorption
capacity towards CO2. In fact, the isotherms even cross at around 18 bar.
When the amount adsorbed is compared not by mass of carbon but by volume, a similar
trend is observed. The calculus involved in the determination of the bulk densities for
ANGUARD 6 and ACHM can be consulted in APPENDIX A.10. Figure 4.31 shows that at low
pressures ACHM adsorbs more CO2 than ANGUARD 6, but around 29 bar, where the
adsorption isotherms cross, the situation reverses.
0
2
4
6
8
10
12
14
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
64
Figure 4.31 - Single-component adsorption isotherms for of CO2 on ANGUARD 6 and CO2 on ACHM 6 at 303.15 K. The amount adsorbed is represented in moles of carbon dioxide by volume of carbon sample.
The isotherm of CO2 on the honeycomb monolith is much squarer than the ANGUARD
6 isotherm. The isotherm shapes is very important when designing a separation process as
Pressure Swing Adsorption (PSA). PSA is widely study adsorption-based process [62] in which
the adsorbed species are desorbed by decreasing the system pressure [14], [28]. Therefore, if
the isotherm presents a square shape an important amount of energy must be spent in order to
decrease the pressure enough to ensure desorption of the retained compound.
4.4. Summary
In this chapter adsorption equilibrium of carbon dioxide (CO2), nitrogen (N2) and butane
(C4H10) at 303.15K, 323.15K and 353.15K and helium (He) pycnometry at 353.15K, in a range of
0 – 35 bar, on ANGUARD 6 activated carbon was presented. Adsorption equilibrium of CO2 on
an activated carbon honeycomb monolith, at the same temperatures and pressure range, was
also disclosed.
The experimental adsorption equilibrium results obtained showed that all the adsorption
isotherms obtained can be classified as IUPAC Type I, characteristic from microporous
adsorbents.
The adsorption experimental data from each pure gas was fitted employing the Sips
isotherm model. This fitting allowed the estimation of the Sips model parameters. With these
parameters the fractional loading, and the isosteric heat of adsorption, could be
calculated.
The isosteric heats of adsorption as a function of the fractional loading and pressure
were also determined. The isosteric heat decreases with the increase of the fractional loading
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
0.0035
0.004
0.0045
0.005
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
65
and with the increasing of pressure for all the adsorbate species studied. In case of CO2, this
dependence is almost nil due to the values of parameter .
For ANGUARD 6, the higher estimated from the Sips model corresponded to carbon
dioxide. The obtained for CO2 on ANGUARD 6 was 21.66 mol/kg while the obtained for
CO2 on the honeycomb monolith was only 11.80 mol/kg. This indicates that the ANGUARD 6
can adsorb more CO2 than the ACHM for sufficiently high pressures. This fact was also
confirmed by comparison of isotherms at the same temperature (303.15 K). Although the ACHM
adsorbs more CO2 at low pressures (above 18 bar when compared by mass and above 29 bar
when compared by volume) the isotherms of CO2 on both carbons cross, and ANGUARD 6 is
able to adsorb more CO2. Also the isotherm of CO2 on the honeycomb monolith is much
squarer than the ANGUARD 6 isotherm, which indicates that to ensure desorption of the
retained compound, a great amount of energy must be spent to decrease the pressure.
66
67
Chapter 5
5. Conclusions and Suggestions for Future Work
5.1. Conclusions
The work developed within this thesis was divided in two parts:
a) Adsorbent characterization of two activated carbons: ANGUARD 6, which is an
activated carbon in the form of extrudates, supplied by Sutcliffe Speakman
Carbons Ltd. (UK) and an honeycomb monolith from Mast Carbon International
Limited (UK);
b) Single-component adsorption equilibrium measurements of pure gases (carbon
dioxide, nitrogen and butane), performed in a magnetic suspension microbalance,
by the gravimetric method.
To characterize the activated carbons several methods were used. Surface chemistry of
both carbons was analyzed using the Point of Zero Change and Bohem tiration methods. The
results obtained indicated that both adsorbents have an amphoteric surface, characterized by
basic groups (chromene, ketone and pyrones) and by acidic groups (carboxylic groups,
phenols, but not by lactonic groups).
Thermogravimetric analysis (TGA) of the ANGUARD 6 sample was also performed. The
TGA curve showed that the weight of ANGUARD 6 decreases steeply at 50ºC (323.15K), losing
at least 3% of its weight. From 50ºC (323.15K) to 550ºC (823.15K) the weight decreases slowly
and above 600ºC (873.15K), the sample starts to decompose.
Adsorption of nitrogen at 77K was also measured for ANGUARD 6 (at this point the
purchased honeycomb monolith was not yet available). From the adsorption isotherm several
physical proprieties of the adsorbent were determined. The BET Surface Area method
determined that ANGUARD 6 has an internal surface area of 1699.79 m²/g. The t-plot method
gave an external surface area of 52.50 m²/g and a micropore surface area of 1647.29 m²/g.
The Horvath-Kawazoe (HK) method ensured the determination of the pore size distribution for
the microporous adsorbent, indicating that ANGUARD 6 is mostly comprised by micropores but
also have a small amount of mesopores, with sizes slightly above 20 . Also, this method
allowed the determination of the pore volume, used in the determination of the absolute amount
68
adsorbed ( = 0.98 cm3/g). The Density Functional Theory method confirmed the pore size
distribution obtained by HK method.
Mercury porosimetry was also performed for ANGUARD 6. The volume occupied by the
mesopores and macropores were determined (0.87 cm3/g). Combining this volume, with the
micropore volume, the total pore volume ANGUARD 6 was determined ( = 1.85 cm3/g).
The second part of this work consisted in the adsorption equilibrium measurements of
three gases: carbon dioxide (CO2), nitrogen (N2) and butane (C4H10) on ANGUARD 6.
Adsorption equilibrium of CO2 on the honeycomb monolith was also measured. The
experiments were made at 303.15K, 323.15K and 353.15K in a pressure range of 0-35 bar. The
experiments were performed by the gravimetric method, using a magnetic suspension
microbalance from Rubotherm GmbH.
The net, excess and total amount adsorbed were calculated to each pure gas. The
isotherms obtained can be classified as IUPAC Type I, typical for microporous adsorbents. The
total amount adsorbed was fitted with the Sips isotherm model. The fitting obtained were good
and allowed the determination of the Sips model parameters.
The isosteric heat was also studied for all the adsorbates studied. The isosteric heat
decreased with the fractional loading and with pressure for all the species studied.
ANGUARD 6 presents a high surface area and micropore volume. This adsorbent also
demonstrated to have a good adsorption capacity towards CO2 and proved to be selective for
CO2/N2 separation. Therefore, ANGUARD 6 can be envisioned as a potential adsorbent to be
employed in the capture of CO2 from flue gases emitted from fossil fueled power stations. More
specifically, ANGUARD 6 can be a good candidate to be used in PSA technology as an
alternative to amine scrubbing in the post-combustion CO2 capture process.
69
5.2. Suggestions for Future Work
Despite the work already developed and disclosed in the pages of this dissertation,
there is still work to be developed regarding the measurement of adsorption properties of the
materials employed.
The activated carbon honeycomb monolith was only available in the final part of this
work and, therefore, its characterization could not be concluded. Thus, its characterization
should be concluded in order to evaluate properties as its pore volume, surface area and pore
size distribution.
Also, measurement of the adsorption equilibrium of methane (CH4) on both adsorbents
would be interesting. Biogas is, nowadays, an interesting energy source and since biogas is
mainly constituted by CH4 and CO2 it would be interesting to evaluate the adsorption properties
of both adsorbents towards CH4. This way, the potential use of the adsorbents on biogas
upgrading can be evaluated.
Finally, determination of the adsorption kinetics of the pure gases in the two activated
carbons should be accomplished. The design of adsorption separation processes depends not
only on the knowledge of the adsorption equilibrium, but also depends strongly on its kinetics.
For this reason, the study of the kinetics of CO2, N2 and CH4 on ANGUARD 6 and the ACHM
should be performed in the near future.
70
71
References
[1] Shen C., Grande C. A., Li P., Yu J., Rodrigues A. E., "Adsorption equilibria and kinetics of
CO2 and N2 in activated carbon beads," Chemical Engineering Journal, vol. 160, pp. 398-
407, 2010.
[2] EPA - United States Environmental Protection Agency, "Overview of Greenhouse Gases,"
27th January 2014. [Online]. Available: http://www.epa.gov/. [Accessed 2014 March 3rd].
[3] European Comission , "Energy Efficiency," 2014. [Online]. Available: http://ec.europa.eu/.
[Accessed 2014 March 15th].
[4] Aker Solutions, "Aker Solutions to perform world's first tests for capture of CO2 from
cement industry," 2014. [Online]. Available: http://www.akersolutions.com/. [Accessed 15
March 2014].
[5] ICO2N, "What is CCS," 2014. [Online]. Available: http://www.ico2n.com/. [Accessed 5th
March 2014].
[6] Bellona Environmental CCS Team, "Carbon Capture and Storage," 2014. [Online].
Available: http://bellona.org/ccs/technology/capture/pre-combustion. [Accessed 5th March
2014].
[7] Gregory L. B., Scharmann W. G., "Carbon Dioxide Scrubbing by Amine Solutions -
Adsorption and Extraction Symposium," Ind. Eng. Chem, p. 514–519, May 1937.
[8] Ribeiro, R. P. P. L., "Electric Swing Adsorption for Gas Separation and Purification," Porto,
2013.
[9] Kutz M., "Clean Power Generation from Coal - CO2 Capture," in Environmentally
Conscious Alternative Energy Production, John Wiley & Sons, 2007, p. (255) 308.
[10] Ribeiro R. P., Sauer T. P., Lopes F. V., Moreira R. F., Grande C. A., Rodrigues A. E.,
"Adsorption of CO2, CH4 and N2 in Activated Carbon Honeycomb Monolith," J. Chem.
Eng., vol. 53, pp. 2311-2317, 9th April 2008.
[11] Condon, James B., Surface Area and Porosity by Physisorption: Measurements and
Theory, Roane State Community College: Elsevier, 2006.
[12] Olds W., Xue Y.,, "Remediation of PAH Contaminated Soils and Groundwater using
Activated Carbon," no. ENNR 429: Midyear Report, 2009.
[13] Rouquerol F., Rouquerol J., Sing K., Adsorption by Powders and Porous Solids: Principles,
Methodology and Applications, San Diego, USA: Academic Press, 1999.
[14] Ruthven, D. M., Principles of Adsorption and Adsorption Processes, New York: John Wiley
& Sons, Inc., 1984.
[15] Esteves, I. A. A. C., "Gas Separation Processes by Integrated Adsorption and Permeation
Technologies," Lisboa, 2005.
72
[16] Lowell S., Shields J. E., Thomas M. A., Matthias T., Characterization of Porous Solids and
Powders: Surface Area, Pore Size and Density, Particle Technology Series, pp. 145-148.
[17] Gumma S., Talu O., "Net Adsorption: A Thermodynamic Framework for Supercritical Gas
Adsorption and Storage in Porous Solids," vol. 26, pp. 17013-17023, 2010.
[18] Ullmann's Encyclopedia of Industrial Chemistry, Seventh ed., John Wiley & Sons, Inc..
[19] United Nations Framework Convention on Climate Change, "Kyoto Protocol," 2014.
[Online]. Available: http://unfccc.int/kyoto_protocol/. [Accessed 5th March 2014].
[20] "CO2Now.org," CO2Now.org, 2014. [Online]. Available: http://co2now.org/. [Accessed 5th
March 2014].
[21] Encyclopedia Britannica , "Butane," 2014. [Online]. Available: http://www.britannica.com/.
[Accessed 2nd March 2014].
[22] Esteves, I.A.A.C. et al., "Adsorption of natural gas and biogas components on activated
carbon," 2008.
[23] Rouquerol J. et al., "Recommendations for the characterization of porous solids," Pure &
Appl. Chern., Great Britain, 1994.
[24] Grande, C. A., "Biogas Upgrading by Pressure Swing Adsorption, Biofuel's Engineering
Process Technology," 2011.
[25] Keller J.U., Staudt R.,, Gas Adsorption Equilibria: Experimental Methods and Adsorptive
Isotherms, Germany: Springer Science + Business Media, Inc., 2005.
[26] Nagano, S.; Tamon, H.; Adzumi, T.; Nakagawa, K.; Suzuki, T., vol. 38, Carbon, 2000, p.
915.
[27] Coulson J. M., Richardson J. F., Tecnologia Química, vol. III, Fundação Calouste
Gulbenkian, 1982.
[28] Yang, R. T., Adsorbents: Fundamentals and Applications, Hoboken, New Jersey: John
Wiley & Sons, Inc., 2003.
[29] Bell, R.G., "Zeolites - British Zeolite Association," British Zeolite Association, May 2001.
[Online]. Available: http://bza.org. [Accessed 26th November 2013].
[30] Lyubchyk A., "Gas Adsorption in the MIL-53(Al) Metal Organic Framework. Experiments
and Molecular Simulation.," 2013.
[31] MOF technologies, "MOF technologies," MOF Technologies Ltd, 2012. [Online]. Available:
http://moftechnologies.com/. [Accessed 26th November 2013].
[32] Lyubchyk A., Esteves I. A. A. C. , Cruz F. J. A. L. , Mota J. P. B., "Experimental and
Theoretical Studies of Supercritical Methane Adsorption in the MIL-53(Al) Metal Organic
Framework," The Journal of Physical Chemistry C, vol. 115, p. 20628 – 20638, 2011.
[33] Ribeiro, A.M.; Campo, M.C.; Narin, G.; Santos, J.C.; Ferreira, A.; Chang, J.-S.; Hwang,
Y.K.; Seo, Y.-K.; Lee, U.H.; Loureiro, J.M., et al., "Pressure Swing Adsorption Process for
73
the Separation of Nitrogen and Propylene with a MOF Adsorbent MIL-100(Fe)," Separation
and Purification Technology, vol. 110, pp. 101-111, 2013.
[34] Cameron Carbon Incorporated, "Activated Carbon: Manufacture, Structure & Properties,"
Cameron Carbon Incorporated, 2006. [Online]. Available: http://cameroncarbon.com/..
[Accessed 26th November 2013].
[35] Ribeiro A. M., Loureiro J. M., "Breakthrough behaviour of water vapor on activated carbon
filters," pp. 357-360, 2006.
[36] Do. D. D., Adsorption Analysis: Equilibria and Kinetics, vol. 2, Queenslan: Imperial College
Press, 1998.
[37] Gorgulho H. F., et al, "Adsorção de fenol sobre carvão activado em meio alcalino," vol. 29,
pp. 1226-1232, 2006.
[38] Carbon Activated Corp., "Carbon Activated Corp.," Carbon Activated Corp., 2013. [Online].
Available: http://activatedcarbon.com/. [Accessed 26th November 2013].
[39] DESOTEC Activated Carbon, "DESOTEC Activated Carbon," DESOTEC Activated Carbon,
2013. [Online]. Available: http://desotec.com/. [Accessed 26th November 2013].
[40] Thomas Publishing Company, "Producing Activated Carbon," Thomas Publishing
Company, 2013. [Online]. Available: http://thomasnet.com/. [Accessed 26th November
2013].
[41] Sun J., Brady T. A., Rood M. J., Rostam-Abadi M., Lizzio A. A., "Adsorbed natural gas
storage with activated carbon," pp. 246-250.
[42] Sushrut Chemicals, "Activated Carbon," 2006. [Online]. Available:
http://sushrutchemicals.com/.. [Accessed 3rd December 2013].
[43] Ekpete O.A., Horsfall M. JNR, "Preparation and Characterization of Activated Carbon
derived from Fluted Pumpkin Stem Waste (Telfairia occidentalis Hook F)," vol. 1, 2011.
[44] Chemviron Carbon, "Activated Carbon as a Catalyst or for Catalyst Support," 2014.
[Online]. Available: http://www.chemvironcarbon.com/. [Accessed 7th March 2014].
[45] WebMD, LLC., "Activated Charcoal," 2013. [Online]. [Accessed 3rd December 2013].
[46] Lozano D., De la Casa M.A. ,Alcañiz J., Cazorla D., Linares A., "Advances in the study of
methane storage in porous carbonaceous materials," Fuel, vol. 81, pp. 1777-1803, 2002.
[47] CABOT INC., "Cabot Norit Activated Carbon products," 2013. [Online]. [Accessed 3rd
December 2013].
[48] Boehm, H. P., "Surface oxides on carbon and their analysis: a critical assessment," pp.
145-149, 2001.
[49] Shrestha Rajeshwar M., Yadav Amar P., Pokharel Bhadra P., Pradhananga Raja Ram´,
"Preparation and characterization of activated carbon from lapsi (choerospondias axillaris)
seed stone by chemical activation with phosphoric acid," Research Journal of Chemical
74
Sciences, vol. 2, pp. 80-86, October 2012.
[50] Figueiredo J. L., Ramôa Ribeiro F., Catálise Heterogénea, Lisboa: Fundação Calouste
Gulbenkien, 1987.
[51] SING, K. S. W. et al., "Reporting Physisorption Data for Gas/solid systems with Special
Reference to the Determination of Surface Area and Porosity," Pure & App. Chem., Great
Britain, 1985.
[52] Aligizaki K.K., Pore Structure of Cement-Based Materials: Testing, Interpretation and
Requirements, CRC Press, 2006 .
[53] Kurdi J., Tremblay A.Y., "The determination of interaction parameters in the
characterization of polyetherimide gas separation membranes using the Horvath-Kawazoe
model," Desalination , vol. 148, p. 341–346, 8 April 2002.
[54] Horvath G, Kawazoe K., "Method for the calculation of effective pore size distribution in
molecular sieve carbon," Journal of Chemical Engineering of Japan, 1983.
[55] Yang R. T, Rege S. U.,, "Corrected Horváth-Kawazoe Equations for Pore-Size Distribution,"
AIChE Journal, vol. 46, April 2000.
[56] Cuevas, J. C., "www.uam.es," Institut für Theoretische Festkörpephysik Universität
Karlsruhe (Germany). [Online]. [Accessed 17th December 2013].
[57] Micromeritics, "Porosimetry," One Micromeritics Drive, 1st January 2001. [Online].
Available: www.micromeritics.com. [Accessed 16th December 2013].
[58] Webb P. A.,, "Volume and Density Determinations for Particle Technologists," February
2001.
[59] Eusébio, M. F. J., Development of an universal interface for monitoring and control of
chemical and biochemical processes, Faculdade de Ciências e Tecnologia da Universidade
Nova de Lisboa (FCT/UNL), Lisbon, 2006.
[60] RUBOTHERM PRÄZIONSMESSTECHKIN GmBH, SORPTION - ISOSORP - THE NEW
SORPTION SUSPENSION BALANCE, p. 8.
[61] National Institute of Standards and Technology, "NIST Chemistry WebBook," 2014.
[Online]. Available: http://webbook.nist.gov/chemistry/. [Accessed 2014].
[62] Ribeiro, A. M.; Santos, J.; Rodrigues, A., , "Pressure Swing Adsorption for CO2 Capture in
Fischer-Tropsch Fuels Production from Biomass," Adsorption-Journal of the International
Adsorption Society , Vols. 17, (3), pp. 443-452, 2011.
[63] Goertzen S. L., Thériault K. D., Oickle A. M., Tarasuk A. C., "Standardization of the Bohem
titration. Part I. CO2 expulsion and endpoint determination," CARBON, pp. 1252-1261,
2010.
[64] Goertzen S. L., Thériault K. D., Oickle A. M., Tarasuk A. C., "Standardization of the Boehm
titration: Part II. Method of agitation, effect of filtering and dilute titrant," CARBON, p. 3313 –
3322, 2010.
75
APPENDIX
76
A. Results from Chapter 3
A.1. Calculus used in the analysis of Bohem Titrations Results
Calculus:
The determination of the number of moles of the surface functionalities (surface
functional groups) at the carbon surface was obtained through the following equation [63]:
(Equation A.1)
Where, is the number of moles of carbon surface functionalities at the carbon
surface; and are the concentration and volume of the basic solution (or acid in case of
HCl) mixed with carbon. The product between the two is the number of moles of the basic
solution (or acid in case of HCl) that will be available to react with the surface groups at the
carbon surface; is the volume of the aliquot taken from ; and (or in case of HCl,
and ) are the concentration and volume of the titrant added to the aliquot and the
product is the number of moles acid (or basic) added to the aliquot and available to react with
the remaining reaction base (or acid);
The quantification of the functional groups at the carbon surface can be calculated
through several subtractions. The NaOH neutralizes all the acid groups (phenols, lactonic and
carboxylic groups) and therefore has a that includes all of these groups. Na2CO3 reacts
with carboxylic and lactonic groups and the difference between (NaOH) and (Na2CO3)
will denote the number of phenols on the surface.
In the same way, since NaHCO3 reacts only with the carboxylic groups the difference
between the (Na2CO3) and (NaHCO3) gives the quantity of lactonic groups. The
quantity of carboxylic groups it is determined directly from (NaHCO3).The quantification of
the basic groups comes from the value of (HCl) [63], [64].
In the case of the ACHM, like it was mention before, the amount available to perform
the experiments was less than for ANGUARD 6, and for this reason only one run, RUN D, was
performed. The same conditions as the ones used in RUN B were employed. Instead of 1.0 g of
well-crushed activated carbon and 10 mL of each solution previously prepared, the quantities
were reduced to half.
77
A.2. Results from Bohem Titrations
Table A.1 - Results from Bohem Titrations Experiments for RUN A – 48 hours without centrifugation (ANGUARD 6).
Solutions Concentration
(M)
pH of the aqueous solution
pH of the solution
mixed with ANG 6
Volume of the
titrated
Weight of ANG
6
pH of the titrant
Volume of the titrant
Concentration of the titrant
Concentration of the titrated
Concentration of the titrated
(0.05M) (292.15K)
(titrated) (292.35K)
(mL) (g) (292.25K) (mL) (mol/mL) (mol/mL) (mmol/mL)
NaOH 0.05 12.15 9.91 5.1 1.015 1.71 1.5 1.000E-04 2.941E-05 0.029
NaHCO3 0.05 8.32 9.01 3.8 1.007 1.71 0.9 1.000E-04 2.368E-05 0.024
Na2CO3 0.05 10.89 9.50 4.0 1.009 1.71 1.4 1.000E-04 3.500E-05 0.035
HCL 0.05 2.07 5.04 2.9 1.003 12.65 0.4 1.000E-04 1.379E-05 0.014
Table A.2 - Results from Bohem Titrations Experiments for RUN B – 48 hours with centrifugation (ANGUARD 6).
Solutions Concentration
(M)
pH of the aqueous solution
pH of the solution
mixed with ANG 6
Volume of the
titrated
Weight of ANG
6
pH of the titrant
Volume of the titrant
Concentration of the titrant
Concentration of the titrated
Concentration of the titrated
(0.05M) (292.45K)
(titrated) (293.15K)
(mL) (g) (292.25) (mL) (mol/mL) (mol/mL) (mmol/mL)
NaOH 0.05 12.15 9.96 6.1 1.003 1.71 1.7 1.000E-04 2.787E-05 0.028
NaHCO3 0.05 8.32 9.40 6.8 1.01 1.71 0.9 1.000E-04 1.324E-05 0.013
Na2CO3 0.05 10.89 9.67 6.8 1.014 1.71 3.1 1.000E-04 4.559E-05 0.046
HCL 0.05 2.07 5.12 5.5 1.006 12.65 0.3 1.000E-04 5.455E-06 0.005
78
Table A.3 - Results from Bohem Titrations Experiments for RUN C– 24 hours with centrifugation (ANGUARD 6).
Solutions Concentration
(M)
pH of the aqueous solution
pH of the solution
mixed with ANG 6
Volume of the
titrated
Weight of ANG
6
pH of the titrant
Volume of the titrant
Concentration of the titrant
Concentration of the titrated
Concentration of the titrated
(0.05M) (292.45K)
(titrated) (293.15K)
(mL) (g) (292.25K) (mL) (mol/mL) (mol/mL) (mmol/mL)
NaOH 0.05 12.15 10.960 6.5 1.011 1.71 2.0 1.000E-04 3.077E-05 0.031
NaHCO3 0.05 8.32 9.800 4.8 1.022 1.71 1.1 1.000E-04 2.292E-05 0.023
Na2CO3 0.05 10.89 10.380 6.5 1.001 1.71 3.2 1.000E-04 4.923E-05 0.049
HCL 0.05 2.07 3.840 5.3 1.000 12.65 0.6 1.000E-04 1.132E-05 0.011
Table A.4 - Results from Bohem Titrations Experiments for RUN D – 48 hours with centrifugation (ACHM).
Solutions Concentration
(M)
pH of the aqueous solution
pH of the solution
mixed with ACHM
Volume of the
titrated
Weight of
ACHM
pH of the titrant
Volume of the titrant
Concentration of the titrant
Concentration of the titrated
Concentration of the titrated
(0.05M) (292.45K)
(titrated) (293.15K)
(mL) (g) (293.15K) (mL) (mol/mL) (mol/mL) (mmol/mL)
NaOH 0.05 12.83 9.61 2.6 0.491 1.02 1.0 1.000E-04 3.8467E-05 0.038
NaHCO3 0.05 9.01 9.87 3.0 0.504 1.02 1.1 1.000E-04 3.667E-05 0.037
Na2CO3 0.05 11.67 10.13 3.0 0.505 1.02 1.4 1.000E-04 4.667E-05 0.047
HCL 0.05 1.48 3.58 2.0 0.493 12.78 0.4 1.000E-04 2.000E-05 0.020
79
A.3. Results from N2 adsorption at 77K
Table A.5 - N2 adsorption isotherm at 77K for ANGUARD 6.
Relative Pressure (mmHg) Absolute Pressure (mmHg) Amount adsorbed (cm3/g STP) Saturation Pressure (mmHg)
5.40E-06 0.0042 20.1990 773.3922
9.80E-06 0.0076 40.3963
2.10E-05 0.0162 60.5900
3.99E-05 0.0309 80.7795
6.86E-05 0.0530 100.9655
1.13E-04 0.0871 121.1435
1.82E-04 0.1403 141.3116
2.95E-04 0.2276 161.4585
3.73E-04 0.2876 171.1009
7.47E-04 0.5761 197.3558
1.50E-03 1.1537 220.5257
3.00E-03 2.3143 241.1775
6.09E-03 4.6931 261.2856
1.22E-02 9.4126 282.0022 770.5537
2.51E-02 19.3363 307.5154
4.89E-02 37.7144 339.4128
9.83E-02 75.7452 390.3850
1.48E-01 113.9255 434.0811
1.99E-01 153.6747 475.4101
2.53E-01 194.8087 513.7624
3.09E-01 238.1761 548.4652
4.04E-01 311.7136 592.2765
5.04E-01 388.4846 618.3242
6.13E-01 472.3442 632.9579
7.09E-01 546.3831 640.0570
7.94E-01 612.1602 645.0776
8.58E-01 661.6496 648.7496
8.99E-01 693.1236 651.4601
9.24E-01 712.8140 653.5630
9.49E-01 731.7781 656.3436
9.73E-01 750.1361 661.1551
9.88E-01 762.0043 667.5474
8.96E-01 691.0161 656.5646
7.89E-01 608.5940 650.5465
6.88E-01 530.6953 645.2773
6.01E-01 463.4106 640.2210 771.2968
5.01E-01 386.5867 633.1470
3.97E-01 306.3074 591.6652
3.03E-01 233.4853 545.7287
2.00E-01 153.9131 475.8777
1.02E-01 78.7608 393.3008
80
A.4. Results from BET Surface Area Method Analysis
Figure A.1- BET Surface Area Report and BET Surface Area Plot for ANGUARD 6, obtained from
DataMasterTM
, V4.00 (2004). The value of is indicated as Qm.
A.5. Results from t-Plot Method Analysis
Figure A.2 - t-Plot Report and t-Plot for ANGUARD 6, obtained from DataMasterTM
, V4.00 (2004).
81
A.6. Results from Horvath-Kawazoe (HK) Method Analysis
Figure A.3 - Horvath-Kawazoe Report for ANGUARD 6, obtained from DataMasterTM
, V4.00 (2004).
A.7. Results from Density Functional Theory (DFT) Method
Analysis
Figure A.4 - Density Functional Theory results for ANGUARD 6, obtained from DataMasterTM
, V4.00 (2004).
A.8. Results from Mercury Porosimetry Analysis
Figure A.5 - Intrusion Data from Hg porosimetry for ANGUARD 6
82
A.9. Resume of the physical parameters calculated from the
several characterization methods for ANGUARD 6
Table A.6 - Characterization physical parameters of ANGUARD 6.
A.10. Calculus used for the determination of the bulk densities
for ANGUARD 6 and ACHM
Bulk density of ANGUARD 6
In order to determine the bulk density of ANGUARD 6, a graduated cylinder of 50 mL
1.0 mL was filled with the carbon pellets. The weights of the graduated cylinder and the
graduated cylinder + the carbon pellets were measured. The weight of the carbon pellets was
estimated by the difference.
= 93.868 g
= 111.909 g
= 18.041 g
83
The bulk density was than calculated by the quotient between the weight of the carbon
pellets and the volume of the graduated cylinder.
Bulk density of ACHM
Since the honeycomb monolith has a cylinder shape, it was easy to determine its
volume. Figure A.6 illustrates a piece of the ACHM and its dimensions.
D = 3 cm
H = 2 cm
Figure A.6 - Illustration of a piece of the ACHM used for the determination of its bulk density.
After the dimensions were recorded, the piece of the ACHM was weighed.
= 5.997 g
= 7.069 cm2
= 2 cm
14.137 cm3
84
85
B. Results from Chapter 4
Table B.1 - Experimental data obtained from helium measurements at 353.15K for ANGUARD 6 and ACHM.
He adsorption at 353.15K for ANG6
He adsorption at 353.15K for ACHM
Weight (g)
ρg (kg/m3)
Weight (g)
ρg (kg/m
3)
6.9016
0.0000
5.9985
0.0000
6.9016
0.0685
5.9985
0.0685
6.9014
0.2928
5.9983
0.2928
6.9007
0.9491
5.9977
0.9491
6.9000
1.8180
5.9972
1.8180
6.8985
3.3762
5.9958
3.3762
6.8976
4.1946
5.9950
4.1946
6.8970
4.6663
5.9945
4.6663
6.8992
2.5652
5.9964
2.5652
6.9006
1.3359
5.9976
1.3359
6.9013
0.5476
5.9983
0.5476
6.9019
0.0000
5.9985
0.0000
86
Table B.2 – Experimental nitrogen adsorption equilibrium data on the carbon sample ANGUARD 6 at 303.15K, 323.15K and 353.15K. 54 experimental data points were measured.
303.15K 323.15K 353.15K
P (bar)
(mol/kg)
(mol/kg)
(mol/kg) P (bar)
(mol/kg)
(mol/kg)
(mol/kg)
P (bar)
(mol/kg)
(mol/kg)
(mol/kg)
0.0048 0.0017 0.0018 0.0020 0.0052 0.0010 0.0011 0.0013 0.1055 0.0118 0.0132 0.0167
0.0245 0.0051 0.0055 0.0064 0.0176 0.0024 0.0027 0.0033 0.7156 0.0699 0.0792 0.1031
0.0952 0.0224 0.0238 0.0275 0.1014 0.0091 0.0106 0.0143 1.0547 0.1124 0.1261 0.1613
0.3088 0.0796 0.0843 0.0963 0.7544 0.0981 0.1088 0.1363 2.9982 0.2511 0.2900 0.3900
0.7642 0.1985 0.2101 0.2398 1.0172 0.1292 0.1436 0.1807 7.1361 0.4904 0.5829 0.8209
1.0280 0.2535 0.2690 0.3090 3.0153 0.3481 0.3909 0.5008 12.0574 0.7243 0.8806 1.2821
3.0544 0.5489 0.5952 0.7140 5.5639 0.5826 0.6616 0.8646 17.1165 0.9214 1.1431 1.7128
8.0958 1.0962 1.2189 1.5340 10.0009 0.9058 1.0478 1.4126 22.2872 1.0912 1.3799 2.1215
13.0752 1.4886 1.6867 2.1959 15.0174 1.2013 1.4144 1.9621 28.1797 1.2469 1.6112 2.5472
18.1291 1.7957 2.0706 2.7770 20.4011 1.4529 1.7424 2.4861 33.0759 1.3524 1.7795 2.8769
22.9614 2.0304 2.3787 3.2736 24.9887 1.6300 1.9844 2.8952 30.3840 1.2919 1.6845 2.6933
27.9974 2.2265 2.6512 3.7425 30.1453 1.7973 2.2246 3.3227 25.2169 1.1734 1.4996 2.3376
31.8068 2.3479 2.8305 4.0705 34.1373 1.8905 2.3744 3.6176 19.7078 1.0002 1.2553 1.9110
25.5976 2.1450 2.5333 3.5311 25.9843 1.6687 2.0372 2.9841 14.5531 0.8246 1.0133 1.4981
15.3769 1.6448 1.8779 2.4769 17.4601 1.3292 1.5770 2.2136 9.3453 0.5963 0.7175 1.0291
10.4483 1.3143 1.4726 1.8794 12.6342 1.0857 1.2651 1.7259 5.8770 0.4122 0.4885 0.6845
4.0172 0.7067 0.7676 0.9239 7.4972 0.7450 0.8515 1.1249 4.0883 0.3048 0.3579 0.4943
2.0575 0.4556 0.4867 0.5668 3.9986 0.4475 0.5042 0.6501 1.9162 0.1638 0.1887 0.2527
2.0161 0.2508 0.2794 0.3529 0.3994 0.0447 0.0499 0.0633
0.4383 0.0847 0.0909 0.1069
87
Figure B.1 - Net (◊), excess (□) and total (∆) adsorption isotherms of nitrogen on ANGUARD 6 at 303.15K (top), 323.15K (middle) and 353.15K (bottom).
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
N2 adsorption isotherm on ANG6 at 323.15K
qnet qex qt
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0 5 10 15 20 25 30 35
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
N2 adsorption isotherm on ANG6 at 353.15K
qnet qex qt
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0 5 10 15 20 25 30 35
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
N2 adsorption isotherm on ANG6 at 303.15K
qnet qex qt
88
Table B.3 - Experimental butane adsorption equilibrium data on the carbon sample ANGUARD 6 at 303.15K, 323.15K and 353.15K. 40 experimental data points were measured.
303.15K 323.15K 353.15K
P (bar)
(mol/kg)
(mol/kg)
(mol/kg) P (bar)
(mol/kg)
(mol/kg)
(mol/kg)
P (bar)
(mol/kg)
(mol/kg)
(mol/kg)
0.0018 0.6981 0.6982 0.6982 0.0066 0.7175 0.7176 0.7178 0.0364 1.0252 1.0257 1.0270
0.0059 1.2985 1.2986 1.2988 0.0108 0.9571 0.9573 0.9577 0.0822 1.5183 1.5195 1.5223
0.0097 1.6463 1.6465 1.6469 0.0342 1.7557 1.7562 1.7575 0.3914 3.0472 3.0530 3.0661
0.0447 2.9250 2.9257 2.9274 0.0890 2.3123 2.3137 2.3170 0.8674 3.9330 3.9460 3.9754
0.0834 3.5221 3.5234 3.5266 0.2883 3.8531 3.8578 3.8684 2.1419 5.1309 5.1638 5.2380
0.5921 6.0561 6.0652 6.0886 0.7728 5.1751 5.1878 5.2165 4.4309 6.4442 6.5154 6.6758
1.0740 7.3356 7.3523 7.3954 1.0045 5.5960 5.6126 5.6501 8.6204 7.3348 7.4879 7.8328
2.0952 8.1699 8.2036 8.2903 2.2484 7.2691 7.3074 7.3939 5.8598 6.9672 7.0643 7.2833
1.5181 7.8986 7.9226 7.9843 3.0863 7.8173 7.8712 7.9926 3.1090 5.7599 5.8086 5.9183
0.8427 6.7919 6.8050 6.8385 3.9800 8.0334 8.1047 8.2654 1.6199 4.7123 4.7369 4.7925
0.3724 5.2927 5.2984 5.3130 1.6560 6.6846 6.7125 6.7753 0.6134 3.5661 3.5753 3.5960
0.1564 4.2081 4.2105 4.2166 0.5500 4.8201 4.8292 4.8495 0.2426 2.6475 2.6511 2.6592
0.0691 3.4227 3.4237 3.4264 0.1706 3.5435 3.5463 3.5525 0.0174 0.7175 0.7178 0.7184
0.0242 2.4727 2.4731 2.4740 0.0279 1.8670 1.8675 1.8685
0.0036 1.0881 1.0882 1.0883 0.0045 0.7965 0.7965 0.7967
89
Figure B.2 -Net (◊), excess (□) and total (∆) adsorption isotherms of butane on ANGUARD 6 at 303.15K (top), 323.15K (middle) and 353.15K (bottom).
0
1
2
3
4
5
6
7
8
9
0 1 2 3 4 5
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
C4H10 adsorption isotherm on ANG6 at 323.15K
qnet qex qt
0
1
2
3
4
5
6
7
8
9
0 2 4 6 8 10
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
C4H10 adsorption isotherm on ANG6 at 353.15K
qnet qex qt
0
1
2
3
4
5
6
7
8
9
0.0 0.5 1.0 1.5 2.0 2.5
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
C4H10 adsorption isotherm on ANG6 at 303.15K
qnet qex qt
90
Table B.4 - Experimental carbon dioxide adsorption equilibrium data on the carbon sample ANGUARD 6 at 303.15K, 323.15K and 353.15K. 43 experimental data points were measured.
303.15K 323.15K 353.15K
P (bar)
(mol/kg)
(mol/kg)
(mol/kg) P (bar)
(mol/kg)
(mol/kg)
(mol/kg)
P (bar)
(mol/kg)
(mol/kg)
(mol/kg)
0.0445 0.0636 0.0643 0.0660 0.0489 0.0475 0.0482 0.0500 0.0953 0.0360 0.0372 0.0404
0.1261 0.1764 0.1783 0.1832 0.0936 0.0860 0.0873 0.0907 0.3554 0.1568 0.1614 0.1733
0.3130 0.4097 0.4144 0.4266 0.3712 0.3051 0.3104 0.3239 0.6792 0.3452 0.3540 0.3767
0.6816 0.7476 0.7536 0.7692 0.7105 0.5387 0.5488 0.5748 1.0561 0.5083 0.5221 0.5574
1.0210 1.0836 1.0991 1.1390 1.0225 0.7262 0.7408 0.7782 5.1716 1.5869 1.6551 1.8302
5.0240 3.6830 3.7609 3.9610 5.0448 2.5310 2.6040 2.7916 10.2313 2.5888 2.7257 3.0774
9.8602 5.6386 5.7953 6.1978 10.0215 4.0429 4.1909 4.5711 14.8082 3.3328 3.5338 4.0502
15.0124 7.1428 7.3881 8.0185 15.8741 5.3549 5.5953 6.2128 19.7528 3.9822 4.2543 4.9535
19.7043 8.2186 8.5496 9.4001 18.0277 5.7557 6.0312 6.7393 25.4129 4.5930 4.9494 5.8652
24.5207 9.1841 9.6086 10.6993 24.5927 6.7710 7.1585 8.1542 29.8137 4.9968 5.4209 6.5109
30.1224 10.1437 10.6855 12.0775 29.8378 7.4344 7.9169 9.1566 34.1600 5.3505 5.8437 7.1110
35.1388 10.8885 11.5447 13.2308 33.9171 7.8764 8.4367 9.8764 23.2206 4.3756 4.6990 5.5301
13.4264 6.7357 6.9531 7.5119 23.1519 6.5637 6.9260 7.8572 3.1077 1.0706 1.1113 1.2160
3.1748 2.6973 2.7461 2.8714 3.3207 1.8595 1.9073 2.0299 1.5708 0.6278 0.6483 0.7010
1.5765 1.5980 1.6221 1.6838 1.5912 1.0525 1.0752 1.1336
0.0224 0.0626 0.0630 0.0638 0.0360 0.0481 0.0486 0.0500
91
Figure B.3 - Net (◊), excess (□) and total (∆) adsorption isotherms of carbon dioxide on ANGUARD 6 at 303.15K (top), 323.15K (middle) and 353.15K (bottom).
(Blank page. Please do not consider this page)
0
2
4
6
8
10
12
14
0 10 20 30 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
CO2 adsorption isotherm on ANG6 at 303.15K
qnet qex qex
0
2
4
6
8
10
12
0 10 20 30 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
CO2 adsorption isotherm on ANG6 at 323.15K
qnet qex qt
0
1
2
3
4
5
6
7
8
0 10 20 30 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
CO2 adsorption isotherm on ANG6 at 353.15K
qnet qex qt
92
Table B.5 - Experimental carbon dioxide adsorption equilibrium data on ACHM at 303.15K, 323.15K and 353.15K. 41 experimental data points were measured.
303.15K 323.15K 353.15K
P (bar)
(mol/kg)
(mol/kg)
(mol/kg) P (bar)
(mol/kg)
(mol/kg)
(mol/kg)
P (bar)
(mol/kg)
(mol/kg)
(mol/kg)
0.0445 0.2084 0.2090 0.2090 0.0936 0.2832 0.2844 0.2844 0.0953 0.1377 0.1388 0.1388
0.1261 0.5503 0.5520 0.5520 0.3712 0.8957 0.9006 0.9006 0.3554 0.4973 0.5016 0.5016
0.3130 1.1820 1.1864 1.1864 0.7105 1.4487 1.4580 1.4580 0.6792 1.0534 1.0616 1.0616
0.6816 2.0122 2.0178 2.0178 1.0225 1.8759 1.8893 1.8893 1.0561 1.4246 1.4372 1.4372
1.0210 2.6879 2.7022 2.7022 5.0448 4.6886 4.7559 4.7559 5.1716 3.2839 3.3467 3.3467
5.0240 6.0145 6.0862 6.0862 10.0215 6.0383 6.1746 6.1746 10.2313 4.4303 4.5564 4.5564
9.8602 7.2441 7.3884 7.3884 15.8741 6.7796 7.0010 7.0010 14.8082 5.0933 5.2784 5.2784
15.0124 7.7803 8.0062 8.0062 18.0277 6.9377 7.1915 7.1915 19.7528 5.5262 5.7768 5.7768
19.7043 7.9952 8.3001 8.3001 24.5927 7.2193 7.5763 7.5763 25.4129 5.8598 6.1881 6.1881
24.5207 8.0591 8.4501 8.4501 29.8378 7.3046 7.7490 7.7490 29.8137 6.0267 6.4174 6.4174
30.1224 8.0314 8.5304 8.5304 33.9171 7.3298 7.8458 7.8458 34.1600 6.1140 6.5683 6.5683
35.1388 7.9005 8.5049 8.5049 23.1519 7.2130 7.5468 7.5468 23.2206 5.7941 6.0920 6.0920
13.4264 7.7225 7.9228 7.9228 3.3207 3.8471 3.8910 3.8910 3.1077 2.5038 2.5413 2.5413
3.1748 5.1124 5.1573 5.1573 1.5912 2.5311 2.5520 2.5520 1.5708 1.6783 1.6971 1.6971
1.5765 3.6160 3.6381 3.6381 0.0360 0.1290 0.1294 0.1294
93
Figure B.4 - Net (◊), excess (□) and total (∆) adsorption isotherms of carbon dioxide on the ACHM at 303.15K (top), 323.15K (middle) and 353.15K (bottom).
0
2
4
6
8
10
0 5 10 15 20 25 30 35 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
CO2 adsorption isotherm on ACHM at 303.15K
qnet qex qt
0
2
4
6
8
10
0 10 20 30 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
CO2 adsorption isotherm on ACHM at 323.15K
qnet qex qt
0
1
2
3
4
5
6
7
0 10 20 30 40
Am
ou
nt
adso
rbe
d (
mo
l/kg
)
Pressure (bar)
CO2 adsorption isotherm on ACHM at 353.15K
qnet qex qt
94
95
C. Equipment Description
C.1. Equipment Description for Adsorption Equilibrium
High-pressure magnetic suspension balance (MSB)
Model: ISOSORP 2000 coupled with a Sartorius microbalance Model BP211D.
Supplier: RUBOTHERM GmbH.
Characteristics: for a maximum load of 25 g (total measuring volume, ie. suspension
coupling and measuring cell with sample), the balance have a resolution of 0.01 mg, an
uncertainty than 0.002% of the measured value, a reproducibility than 0.03 mg, for
pressures in the range UHV - 150 bar and temperatures up to 100 ºC.
Vacuum Pump (VP)
Model: EDWARDS 5 C, A65201903I
Pump Serial Number: 139482910
Supplier: EDWARDS
Characteristics: pumping speed 3.0 m3/h, motor type: RV3 US/EUR PUMP HIGH VOLTS ,
220-240 V, 50/60 Hz, Single phase, operating temperature of – 30ºC to 70ºC, maximum
total pressure in high flux of 1.2×10−1, maximum total pressure of 2×10−3.
Pump Oil: Edwards Ultragrade 19, hydrocarbon-oil, H11025015, 1 Litre.
Pressure Generator
Model: 87-6-5
Supplier: HiP
Characteristics: pressure rating of 5000 psi, capacity per stroke 60 ml with teflon packing
B-208.
Thermostatic Bath – Refrigerator/Heater
Model: F32-HL
Supplier: JULABO Labortechnik GmbH
Characteristics: working temperature range of -35ºC to 200ºC, temperature stability 0,01
ºC, cooling capacity: +20 0 -20 -30ºC to (Medium: ethanol): 0,45 0,39 0,15 0,06 KW, overall
dimensions 31x42x64 cm, bath opening (WxL) 18x12 cm, bath depth 15 cm, filling volume
5.5 to 8 liters , weight 38 Kg.
96
Refrigerant: R134a
Pt100 Temperature Probes
Model: Pt100
Supplier: RS Amidata, Spain
Characteristics: 4 wires temperature sensors with platinum resistance, that exhibit a
typical resistance of 100 at 0ºC, typically measure temperatures up 850ºC, Classe B
precision ±0.12 at 0.3ºC. It consists of a thin film of platinum on a plastic film inside a
stainless steel involucre. The relationship between resistance and temperature is relatively
linear, but curve fitting is often the most accurate way to make the RTD measurement. The
probes were calibrated in the laboratory against a highly accurate Hart Scientific Pt 5613
temperature sensor with an accuracy of ±0.01 K.
Pressure Transducer (PT)
Model: MKS Baratron Type 627D
Supplier: MKS Baratron
Characteristics: pressure measurements in the range from 1K Torr (1.3157bar) to as low
as 0.02 Torr (0.00002 bar) Full Scale (FS). The instrument operates with 15 VDC ( %5)
input at 250 mA, and provides 0 to 10 VDC output linear with pressure. The 627D
transducer is available with optional heater status LEDs, two interface connector lock
options, and a variety of fittings. The unit is capable of measuring pressure at ambient
temperatures of 15ºC to 40ºC (59ºC to 104ºF).
Pressure Transducer (PT) (OM2)
Model: PX01C1-150A5T
Supplier: OMEGADYNE, Inc.
Characteristics: pressure range of 0-10 bar.
Pressure Transducer (PT) (OM3)
Model: PX01C1-500A5T
Supplier: OMEGADYNE, Inc.
Characteristics: pressure range of 0-35 bar.
97
Pressure Transducer (PT) (OM4)
Model: PX03C1-3KA5T
Supplier: OMEGADYNE, Inc.
Characteristics: pressure range for 0-69 bar-138-207.
Table C.1 - Characteristic of the several pressure transducers used in this work.
Name ACRN. Supply (VDC)
(linearity) ACC
(%F.S.)
F.S. (bar)
Output (VDC)
Ch. Calibration
Y=a+bx
Omega 1
OM1 28 0.005 1.034 0-5 0 USB6 xxx Analog Input Multi Sample
Omega 2
OM2 28 0.005 6.124 0-5 1 USB6 xxx Analog Input Multi Sample
Omega 3
OM3 28 0.005 34.83 0-5 2 USB6 xxx Analog Input Multi Sample
Omega 4
OM4 28 0.15 68.931 0-5 3 USB6 xxx Analog Input Multi Sample
Baratron MKS
15 VDC
5% 250 mA
1.005 0-10 COM
1 MKS PR 4000B
MSB 0-5 COM
4 Sartorius Rubotherm
Power Suppliers Model: PS 613
Supplier: Velleman
Characteristics: variable voltage of 0–30 V, 2.5 A DC and two fixed supplies of
±12 V and ±5 V.
Ball Valves
Model: SS-43S4
Supplier: Swagelok
Characteristics: 1/8”OD fittings, Cv = 2.4, P ≤ 206 bar, 283K ≤ T ≤ 338K
Check Valves
Model: SS-4C-TR-1
Supplier: Swagelok
Characteristics: 1/8”OD fittings, PTFE seals, Pcrack = 0.06 bar, Pmax = 206 bar.
98
Several fittings
Model: Swagelok types (nuts, unions, reducers, elbows, etc.)
Supplier: Swagelok
Characteristics: 1/8”OD fittings.
Computers (PC)
Model: Intel(R) Core(TM) i3-2100 CPU @ 3.10GHz 3.10 GHz
Supplier: Tsunami Computers
Characteristics: Windows 7 Professional, RAM: 8.00 Gb, System type: 64-bit
Operating System.
Gases Supplier: Air Liquide and Praxair (Portugal and Spain).
Characteristics: Compressed Helium (He) (99.99%), P=200 bar from Air Liquide
Alphagaz; Compressed Nitrogen (N2), P=200 bar from Air Liquide Alphagaz; Carbon dioxide
(CO2) N48, P=80 bar from Air Liquide Alphagaz; Butano (C4H10) N35, P=0.75 bar from Air
Liquide Alphagaz.
99
C.2. Bank of Images
Figure C.3 - Unit controller for data acquisition by Rubotherm GmBH.
Figure C.2 - Pressure Transducers from MKS Baratron and
Omegadyne.
Figure C.4 - Gas Bottles from Air Liquid and Praxair.
Figure C.1 - Magnetic
Suspension Balance (Metal version).
100
Figure C.9 - Pictures of the experimental apparatus used in the equilibrium measurements.
Figure C.5 - Pressure Generator from HiP.
Figure C.6 - Thermostatic Bath, Refrigerator/Heater from
Julabo.
Figure C.8 - Heater from Nabertherm. Figure C.7 - Vaccum Pump from Edwards.
101